This manual is for an old version of Hazelcast IMDG, use the latest stable version.
This manual is for an old version of Hazelcast IMDG, use the latest stable version.

Hazelcast Documentation

version 2.3.1

Permission to use, copy, modify and distribute this document for any purpose and without fee is hereby granted in perpetuity, provided that the above copyright notice and this paragraph appear in all copies.

31 August 2012


Table of Contents

1. Introduction
1.1. Getting Started (Tutorial)
2. Distributed Data Structures
2.1. Distributed Queue
2.2. Distributed Topic
2.3. Distributed Map
2.3.1. Backups
2.3.2. Eviction
2.3.3. Persistence
2.3.4. Query
2.3.5. Near Cache
2.3.6. Entry Statistics
2.3.7. Indexing
2.4. Distributed MultiMap
2.5. Distributed Set
2.6. Distributed List
2.7. Distributed Lock
2.8. Distributed Events
3. Elastic Memory (Enterprise Edition Only)
4. Security (Enterprise Edition Only)
4.1. Credentials
4.2. ClusterLoginModule
4.3. Cluster Member Security
4.4. Native Client Security
4.4.1. Authentication
4.4.2. Authorization
4.4.3. Permissions
5. Data Affinity
6. Monitoring with JMX
7. Cluster Utilities
7.1. Cluster Interface
7.2. Cluster-wide Id Generator
7.3. LiteMember
8. Transactions
8.1. Transaction Interface
8.2. J2EE Integration
8.2.1. Resource Adapter Configuration
8.2.2. Sample Glassfish v3 Web Application Configuration
8.2.3. Sample JBoss Web Application Configuration
9. Distributed Executor Service
9.1. Distributed Execution
9.2. Execution Cancellation
9.3. Execution Callback
10. Http Session Clustering with HazelcastWM
11. WAN Replication
12. Configuration
12.1. Creating Separate Clusters
12.2. Network Configuration
12.2.1. Configuring TCP/IP Cluster
12.2.2. Specifying Network Interfaces
12.2.3. EC2 Auto Discovery
12.2.4. Network Partitioning (Split-Brain Syndrome)
12.2.5. SSL
12.2.6. Encryption
12.2.7. Socket Interceptor
12.2.8. IPv6 Support
12.3. Partition Group Configuration
12.4. Listener Configurations
12.5. Wildcard Configuration
12.6. Advanced Configuration Properties
12.7. Logging Configuration
12.8. Setting License Key (Enterprise Edition Only)
13. Hibernate Second Level Cache
14. Spring Integration
14.1. Configuration
14.2. Spring Managed Context
14.3. Spring Cache
14.4. Hibernate 2nd Level Cache Config
14.5. Spring Data - JPA
14.6. Spring Data - MongoDB
15. Clients
15.1. Native Client
15.1.1. Java Client
15.1.2. CSharp Client (Enterprise Edition Only)
15.2. Memcache Client
15.3. Rest Client
16. Internals
16.1. Internals 2: Serialization
16.2. Internals 3: Cluster Membership
16.3. Internals 4: Distributed Map
17. Management Center
17.1. Introduction
17.1.1. Installation
17.1.2. User Administration
17.1.3. Tool Overview
17.2. Maps
17.2.1. Monitoring Maps
17.2.2. Map Browser
17.2.3. Map Configuration
17.3. Queues
17.4. Topics
17.5. Members
17.5.1. Monitoring
17.5.2. Operations
17.6. System Logs
17.7. Scripting
17.8. Time Travel
17.9. Console
18. Miscellaneous
18.1. Common Gotchas
18.2. Testing Cluster
18.3. Planned Features
18.4. Release Notes

List of Tables

12.1. Properties Table

Chapter 1. Introduction

Hazelcast is a clustering and highly scalable data distribution platform for Java. Hazelcast helps architects and developers to easily design and develop faster, highly scalable and reliable applications for their businesses.

  • Distributed implementations of java.util.{Queue, Set, List, Map}

  • Distributed implementation of java.util.concurrent.ExecutorService

  • Distributed implementation of java.util.concurrency.locks.Lock

  • Distributed Topic for publish/subscribe messaging

  • Transaction support and J2EE container integration via JCA

  • Distributed listeners and events

  • Support for cluster info and membership events

  • Dynamic HTTP session clustering

  • Dynamic clustering

  • Dynamic scaling to hundreds of servers

  • Dynamic partitioning with backups

  • Dynamic fail-over

  • Super simple to use; include a single jar

  • Super fast; thousands of operations per sec.

  • Super small; less than a MB

  • Super efficient; very nice to CPU and RAM

To install Hazelcast:

  • Download hazelcast-_version_.zip from www.hazelcast.com

  • Unzip hazelcast-_version_.zip file

  • Add hazelcast.jar file into your classpath

Hazelcast is pure Java. JVMs that are running Hazelcast will dynamically cluster. Although by default Hazelcast will use multicast for discovery, it can also be configured to only use TCP/IP for environments where multicast is not available or preferred (Click here for more info). Communication among cluster members is always TCP/IP with Java NIO beauty. Default configuration comes with 1 backup so if one node fails, no data will be lost. It is as simple as usingjava.util.{Queue, Set, List, Map}. Just add the hazelcast.jar into your classpath and start coding.

1.1. Getting Started (Tutorial)

In this short tutorial, we will create simple Java application using Hazelcast distributed map and queue. Then we will run our application twice to have two nodes (JVMs) clustered and finalize this tutorial with connecting to our cluster from another Java application by using Hazelcast Native Java Client API.

  • Download the latest Hazelcast zip.

  • Unzip it and add the lib/hazelcast.jar to your class path.

  • Create a Java class and import Hazelcast libraries.

  • Following code will start the first node and create and use customers map and queue.

    import com.hazelcast.core.Hazelcast;
    import java.util.Map;
    import java.util.Queue;
    
    public class GettingStarted {
    
        public static void main(String[] args) {
            Map<Integer, String> mapCustomers = Hazelcast.getMap("customers");
            mapCustomers.put(1, "Joe");
            mapCustomers.put(2, "Ali");
            mapCustomers.put(3, "Avi");
    
            System.out.println("Customer with key 1: "+ mapCustomers.get(1));
            System.out.println("Map Size:" + mapCustomers.size());
    
            Queue<String> queueCustomers = Hazelcast.getQueue("customers");
            queueCustomers.offer("Tom");
            queueCustomers.offer("Mary");
            queueCustomers.offer("Jane");
            System.out.println("First customer: " + queueCustomers.poll());
            System.out.println("Second customer: "+ queueCustomers.peek());
            System.out.println("Queue size: " + queueCustomers.size());
        }
    }
    

  • Run this class second time to get the second node started.

  • Have you seen they formed a cluster? You should see something like this:

    Members [2] {
    	Member [127.0.0.1:5701]
    	Member [127.0.0.1:5702] this
    }
                                

Connecting Hazelcast Cluster with Java Client API

  • Besides hazelcast.jar you should also add hazelcast-client.jar to your classpath.

  • Following code will start a Hazelcast Client, connect to our two node cluster and print the size of our customers map.

    package com.hazelcast.test;
    
    import com.hazelcast.client.ClientConfig;
    import com.hazelcast.client.HazelcastClient;
    import com.hazelcast.core.HazelcastInstance;
    import com.hazelcast.core.IMap;
    
    public class GettingStartedClient {
    
        public static void main(String[] args) {
            ClientConfig clientConfig = new ClientConfig();
            clientConfig.addAddress("127.0.0.1:5701");
            HazelcastInstance client = HazelcastClient.newHazelcastClient(clientConfig);
            IMap map = client.getMap("customers");
            System.out.println("Map Size:" + map.size());
        }
    }
    

  • When you run it, you will see the client properly connects to the cluster and print the map size as 3.

What is Next?

  • You can browse documentation and resources for detailed features and examples.

  • You can email your questions to Hazelcast mail group.

  • You can browse Hazelcast source code.

Chapter 2. Distributed Data Structures

Common Features of all Hazelcast Data Structures:

  • Data in the cluster is almost evenly distributed (partitioned) across all nodes. So each node carries ~ (1/n * total-data) + backups , n being the number of nodes in the cluster.

  • If a member goes down, its backup replica that also holds the same data, will dynamically redistribute the data including the ownership and locks on them to remaining live nodes. As a result, no data will get lost.

  • When a new node joins the cluster, new node takes ownership(responsibility) and load of -some- of the entire data in the cluster. Eventually the new node will carry almost (1/n * total-data) + backups and becomes the new partition reducing the load on others.

  • There is no single cluster master or something that can cause single point of failure. Every node in the cluster has equal rights and responsibilities. No-one is superior. And no dependency on external 'server' or 'master' kind of concept.

Here is how you can retrieve existing data structure instances (map, queue, set, lock, topic, etc.) and how you can listen for instance events to get notified when an instance is created or destroyed.

import java.util.Collection;
import com.hazelcast.core.Hazelcast;
import com.hazelcast.core.Instance;
import com.hazelcast.core.InstanceEvent;
import com.hazelcast.core.InstanceListener;

public class Sample implements InstanceListener {
    public static void main(String[] args) {
        Sample sample = new Sample();

        Hazelcast.addInstanceListener(sample);

        Collection<Instance> instances = Hazelcast.getInstances();
        for (Instance instance : instances) {
           System.out.println(instance.getInstanceType() + "," + instance.getId());
        }
    }

    public void instanceCreated(InstanceEvent event) {
       Instance instance = event.getInstance();
       System.out.println("Created " + instance.getInstanceType() + "," + instance.getId());
    }

    public void instanceDestroyed(InstanceEvent event) {
        Instance instance = event.getInstance();
        System.out.println("Destroyed " + instance.getInstanceType() + "," + instance.getId());
    }
}

2.1. Distributed Queue

Hazelcast distributed queue is an implementation ofjava.util.concurrent.BlockingQueue.

import com.hazelcast.core.Hazelcast;
import java.util.concurrent.BlockingQueue;
import java.util.concurrent.TimeUnit;

BlockingQueue<MyTask> q = Hazelcast.getQueue("tasks");
q.put(new MyTask());
MyTask task = q.take();

boolean offered = q.offer(new MyTask(), 10, TimeUnit.SECONDS);
task = q.poll (5, TimeUnit.SECONDS);
if (task != null) {
    //process task
}

If you have 10 million tasks in your "tasks" queue and you are running that code over 10 JVMs (or servers), then each server carries 1 million task objects (plus backups). FIFO ordering will apply to all queue operations cluster-wide. User objects (such as MyTask in the example above), that are (en/de)queued have to beSerializable. Maximum capacity per JVM and the TTL (Time to Live) for a queue can be configured.

Distributed queues are backed by distributed maps. Thus, queues can have custom backup counts and persistent storage. Hazelcast will generate cluster-wide unique id for each element in the queue.

Sample configuration:

<hazelcast>
    ...
    <queue name="tasks">
        <!--
            Maximum size of the queue. When a JVM's local queue size reaches the maximum,
            all put/offer operations will get blocked until the queue size
            of the JVM goes down below the maximum.
            Any integer between 0 and Integer.MAX_VALUE. 0 means Integer.MAX_VALUE. Default is 0.
        -->
        <max-size-per-jvm>10000</max-size-per-jvm>
        
        <!--
            Name of the map configuration that will be used for the backing distributed
            map for this queue.
        -->
        <backing-map-ref>queue-map</backing-map-ref>
    </queue>

    <map name="queue-map">

        <backup-count>1</backup-count>

        <map-store enabled="true">

            <class-name>com.your,company.storage.DBMapStore</class-name>

            <write-delay-seconds>0</write-delay-seconds>

        </map-store>

        ...

    </map>
</hazelcast>

If the backing map has no map-store defined then your distributed queue will be in-memory only. If the backing map has a map-store defined then Hazelcast will call your MapStore implementation to persist queue elements. Even if you reboot your cluster Hazelcast will rebuild the queue with its content. When implementing a MapStore for the backing map, note that type of the key is always Long and values are the elements you place into the queue. So make sure MapStore.loadAllKeys returns Set<Long> for instance.

To learn about wildcard configuration feature, see Wildcard Configuration page.

2.2. Distributed Topic

Hazelcast provides distribution mechanism for publishing messages that are delivered to multiple subscribers which is also known as publish/subscribe (pub/sub) messaging model. Publish and subscriptions are cluster-wide. When a member subscribes for a topic, it is actually registering for messages published by any member in the cluster, including the new members joined after you added the listener. Messages are ordered, meaning, listeners(subscribers) will process the messages in the order they are actually published. If cluster member M publishes messages m1, m2, m3...mn to a topic T, then Hazelcast makes sure that all of the subscribers of topic T will receive and process m1, m2, m3...mn in order. Therefore there is only single thread invoking onMessage. The user shouldn't keep the thread busy and preferably dispatch it via an Executor. This will increase the performance of the topic

import com.hazelcast.core.Topic;
import com.hazelcast.core.Hazelcast;
import com.hazelcast.core.MessageListener;

public class Sample implements MessageListener<MyEvent> {

    public static void main(String[] args) {
        Sample sample = new Sample();
        ITopic topic = Hazelcast.getTopic ("default");
        topic.addMessageListener(sample);
        topic.publish (new MyEvent());
    }

    public void onMessage(MyEvent myEvent) {
        System.out.println("Message received = " + myEvent.toString());
        if (myEvent.isHeavyweight()) {
            messageExecutor.execute(new Runnable() {
                public void run() {
                    doHeavyweightStuff(myEvent);
                }
            });
        }
    }

    // ...

    private static final Executor messageExecutor = Executors.newSingleThreadExecutor();
}

To learn about wildcard configuration feature, see Wildcard Configuration page.

2.3. Distributed Map

Just like queue and set, Hazelcast will partition your map entries; and almost evenly distribute onto all Hazelcast members. Distributed maps have 1 backup (replica-count) by default so that if a member goes down, we don't lose data. Backup operations are synchronous so when a map.put(key, value) returns, it is guaranteed that the entry is replicated to one other node. For the reads, it is also guaranteed that map.get(key) returns the latest value of the entry. Consistency is strictly enforced.

import com.hazelcast.core.Hazelcast;
import java.util.Map;
import java.util.Collection;

Map<String, Customer> mapCustomers = Hazelcast.getMap("customers");
mapCustomers.put("1", new Customer("Joe", "Smith"));
mapCustomers.put("2", new Customer("Ali", "Selam"));
mapCustomers.put("3", new Customer("Avi", "Noyan"));

Collection<Customer> colCustomers = mapCustomers.values();
for (Customer customer : colCustomers) {
    // process customer
}

Hazelcast.getMap() actually returns com.hazelcast.core.IMap which extends java.util.concurrent.ConcurrentMap interface. So methods like ConcurrentMap.putIfAbsent(key,value) and ConcurrentMap.replace(key,value) can be used on distributed map as shown in the example below.

import com.hazelcast.core.Hazelcast;
import java.util.concurrent.ConcurrentMap;

Customer getCustomer (String id) {
    ConcurrentMap<String, Customer> map = Hazelcast.getMap("customers");
    Customer customer = map.get(id);
    if (customer == null) {
        customer = new Customer (id);
        customer = map.putIfAbsent(id, customer);
    }
    return customer;
}               

public boolean updateCustomer (Customer customer) {
    ConcurrentMap<String, Customer> map = Hazelcast.getMap("customers");
    return (map.replace(customer.getId(), customer) != null);            
}
                
public boolean removeCustomer (Customer customer) {
    ConcurrentMap<String, Customer> map = Hazelcast.getMap("customers");
    return map.remove(customer.getId(), customer) );           
}                                  
        

All ConcurrentMap operations such as put and remove might wait if the key is locked by another thread in the local or remote JVM, but they will eventually return with success. ConcurrentMap operations never throwjava.util.ConcurrentModificationException.

Also see:

2.3.1. Backups

Hazelcast will distribute map entries onto multiple JVMs (cluster members). Each JVM holds some portion of the data but we don't want to lose data when a member JVM crashes. To provide data-safety, Hazelcast allows you to specify the number of backup copies you want to have. That way data on a JVM will be copied onto other JVM(s). Hazelcast supports both sync and async backups. Sync backups block operations until backups are successfully copied to backups nodes (or deleted from backup nodes in case of remove) and acknowledgements are received. In contrast, async backups do not block operations, they are fire & forget and do not require acknowledgements. By default, Hazelcast will have one sync backup copy. If backup count >= 1, then each member will carry both owned entries and backup copies of other member(s). So for the map.get(key) call, it is possible that calling member has backup copy of that key but by default, map.get(key) will always read the value from the actual owner of the key for consistency. It is possible to enable backup reads by changing the configuration. Enabling backup reads will give you greater performance.

<hazelcast>
    ...
    <map name="default">
        <!--
            Number of sync-backups. If 1 is set as the backup-count for example,
            then all entries of the map will be copied to another JVM for
            fail-safety. Valid numbers are 0 (no backup), 1, 2, 3.
        -->
        <backup-count>1</backup-count>

        <!--
            Number of async-backups. If 1 is set as the backup-count for example,
            then all entries of the map will be copied to another JVM for
            fail-safety. Valid numbers are 0 (no backup), 1, 2, 3.
        -->
        <async-backup-count>1</async-backup-count>

        <!--
            Can we read the local backup entries? Default value is false for
            strong consistency. Being able to read backup data will give you
            greater performance.
        -->
        <read-backup-data>false</read-backup-data>

        ...
    </map>
</hazelcast>

2.3.2. Eviction

Hazelcast also supports policy based eviction for distributed map. Currently supported eviction policies are LRU (Least Recently Used) and LFU (Least Frequently Used). This feature enables Hazelcast to be used as a distributed cache. If time-to-live-seconds is not 0 then entries older than time-to-live-seconds value will get evicted, regardless of the eviction policy set. Here is a sample configuration for eviction:

<hazelcast>
    ...
    <map name="default">
        <!--
            Number of backups. If 1 is set as the backup-count for example,
            then all entries of the map will be copied to another JVM for
            fail-safety. Valid numbers are 0 (no backup), 1, 2, 3.
        -->
        <backup-count>1</backup-count>

        <!--
            Maximum number of seconds for each entry to stay in the map. Entries that are
            older than <time-to-live-seconds> and not updated for <time-to-live-seconds>
            will get automatically evicted from the map.
            Any integer between 0 and Integer.MAX_VALUE. 0 means infinite. Default is 0.
        -->
        <time-to-live-seconds>0</time-to-live-seconds>

        <!--
            Maximum number of seconds for each entry to stay idle in the map. Entries that are
            idle(not touched) for more than <max-idle-seconds> will get
            automatically evicted from the map.
            Entry is touched if get, put or containsKey is called.
            Any integer between 0 and Integer.MAX_VALUE.
            0 means infinite. Default is 0.
        -->
        <max-idle-seconds>0</max-idle-seconds>

        <!--
            Valid values are:
            NONE (no extra eviction, <time-to-live-seconds> may still apply),
            LRU  (Least Recently Used),
            LFU  (Least Frequently Used).
            NONE is the default.
            Regardless of the eviction policy used, <time-to-live-seconds> will still apply. 
        -->
        <eviction-policy>LRU</eviction-policy>

        <!--
            Maximum size of the map. When max size is reached,
            map is evicted based on the policy defined.
            Any integer between 0 and Integer.MAX_VALUE. 0 means
            Integer.MAX_VALUE. Default is 0.
        -->
        <max-size policy="cluster_wide_map_size">5000</max-size>

        <!--
            When max. size is reached, specified percentage of
            the map will be evicted. Any integer between 0 and 100.
            If 25 is set for example, 25% of the entries will
            get evicted.
        -->
        <eviction-percentage>25</eviction-percentage>
       <!--
            Specifies when eviction will be started. Default value is 3. 
           So every 3 (+up to 5 for performance reasons) seconds 
           eviction will be kicked of. Eviction is costly operation, setting 
           this number too low, can decrease the performance. 
       -->
      <eviction-delay-seconds>3</eviction-delay-seconds>
    </map>
</hazelcast>

Max-Size Policies

There are 5 defined policies can be used in max-size configuration.

  1. cluster_wide_map_size: Cluster-wide total max map size (default policy).

    <max-size policy="cluster_wide_map_size">50000</max-size>

  2. map_size_per_jvm: Max map size per JVM.

    <max-size policy="map_size_per_jvm">5000</max-size>

  3. partitions_wide_map_size: Partitions (default 271) wide max map size.

    <max-size policy="partitions_wide_map_size">27100</max-size>

  4. used_heap_size: Max used heap size in MB (mega-bytes) per JVM.

    <max-size policy="used_heap_size">4096</max-size>

  5. used_heap_percentage: Max used heap size percentage per JVM.

    <max-size policy="used_heap_percentage">75</max-size>

2.3.3. Persistence

Hazelcast allows you to load and store the distributed map entries from/to a persistent datastore such as relational database. If a loader implementation is provided, when get(key) is called, if the map entry doesn't exist in-memory then Hazelcast will call your loader implementation to load the entry from a datastore. If a store implementation is provided, when put(key,value) is called, Hazelcast will call your store implementation to store the entry into a datastore. Hazelcast can call your implementation to store the entries synchronously (write-through) with no-delay or asynchronously (write-behind) with delay and it is defined by the write-delay-seconds value in the configuration.

If it is write-through, when the map.put(key,value) call returns, you can be sure that

  • MapStore.store(key,value) is successfully called so the entry is persisted.

  • In-Memory entry is updated

  • In-Memory backup copies are successfully created on other JVMs (if backup-count is greater than 0)

If it is write-behind, when the map.put(key,value) call returns, you can be sure that

  • In-Memory entry is updated

  • In-Memory backup copies are successfully created on other JVMs (if backup-count is greater than 0)

  • The entry is marked as dirty so that after write-delay-seconds, it can be persisted.

Same behavior goes for the remove(key and MapStore.delete(key). If MapStore throws an exception then the exception will be propagated back to the original put or remove call in the form of RuntimeException. When write-through is used, Hazelcast will call MapStore.store(key,value) and MapStore.delete(key) for each entry update. When write-behind is used, Hazelcast will callMapStore.store(map), and MapStore.delete(collection) to do all writes in a single call. Also note that your MapStore or MapLoader implementation should not use Hazelcast Map/Queue/MultiMap/List/Set operations. Your implementation should only work with your data store. Otherwise you may get into deadlock situations.

Here is a sample configuration:

<hazelcast>
    ...
    <map name="default">
        ...
        <map-store enabled="true">
            <!--
               Name of the class implementing MapLoader and/or MapStore.
               The class should implement at least of these interfaces and
               contain no-argument constructor. Note that the inner classes are not supported.
            -->
            <class-name>com.hazelcast.examples.DummyStore</class-name>
            <!--
               Number of seconds to delay to call the MapStore.store(key, value).
               If the value is zero then it is write-through so MapStore.store(key, value)
               will be called as soon as the entry is updated.
               Otherwise it is write-behind so updates will be stored after write-delay-seconds
               value by calling Hazelcast.storeAll(map). Default value is 0.
            -->
            <write-delay-seconds>0</write-delay-seconds>
        </map-store>
    </map>
</hazelcast>

Initialization on startup:

As of 1.9.3 MapLoader has the new MapLoader.loadAllKeys API. It is used for pre-populating the in-memory map when the map is first touched/used. If MapLoader.loadAllKeys returns NULL then nothing will be loaded. Your MapLoader.loadAllKeys implementation can return all or some of the keys. You may select and return only the hot keys, for instance. Also note that this is the fastest way of pre-populating the map as Hazelcast will optimize the loading process by having each node loading owned portion of the entries.

Here is MapLoader initialization flow;

  1. When getMap() first called from any node, initialization starts

  2. Hazelcast will call MapLoader.loadAllKeys() to get all your keys on each node

  3. Each node will figure out the list of keys it owns

  4. Each node will load all its owned keys by calling MapLoader.loadAll(keys)

  5. Each node puts its owned entries into the map by calling IMap.putTransient(key,value)

2.3.4. Query

Hazelcast partitions your data and spreads across cluster of servers. You can surely iterate over the map entries and look for certain entries you are interested in but this is not very efficient as you will have to bring entire entry set and iterate locally. Instead, Hazelcast allows you to run distributed queries on your distributed map.

Let's say you have a "employee" map containing values of Employee objects:

import java.io.Serializable;

public class Employee implements Serializable {
private String name;
private int age;
private boolean active;
private double salary;

public Employee(String name, int age, boolean live, double price) {
    this.name = name;
    this.age = age;
    this.active = live;
    this.salary = price;
}

public Employee() {
}

public String getName() {
    return name;
}

public int getAge() {
    return age;
}

public double getSalary() {
    return salary;
}

public boolean isActive() {
    return active;
}
}

Now you are looking for the employees who are active and with age less than 30. Hazelcast allows you to find these entries in two different ways:

Distributed SQL Query

SqlPredicate takes regular SQL where clause. Here is an example:

import com.hazelcast.core.IMap;
import com.hazelcast.query.SqlPredicate;

IMap map = Hazelcast.getMap("employee");

Set<Employee> employees = (Set<Employee>) map.values(new SqlPredicate("active AND age < 30"));

Supported SQL syntax:

  • AND/OR

    • <expression> AND <expression> AND <expression>...

      • active AND age>30

      • active=false OR age = 45 OR name = 'Joe'

      • active AND (age >20 OR salary < 60000)

  • =, !=, <, <=, >, >=

    • <expression> = value

      • age <= 30

      • name ="Joe"

      • salary != 50000

  • BETWEEN

    • <attribute> [NOT] BETWEEN <value1> AND <value2>

      • age BETWEEN 20 AND 33 (same as age >=20 AND age<=33)

      • age NOT BETWEEN 30 AND 40 (same as age <30 OR age>40)

  • LIKE

    • <attribute> [NOT] LIKE 'expression'

      % (percentage sign) is placeholder for many characters, _ (underscore) is placeholder for only one character.

      • name LIKE 'Jo%' (true for 'Joe', 'Josh', 'Joseph' etc.)

      • name LIKE 'Jo_' (true for 'Joe'; false for 'Josh')

      • name NOT LIKE 'Jo_' (true for 'Josh'; false for 'Joe')

      • name LIKE 'J_s%' (true for 'Josh', 'Joseph'; false 'John', 'Joe')

  • IN

    • <attribute> [NOT] IN (val1, val2, ...)

      • age IN (20, 30, 40)

      • age NOT IN (60, 70)

Examples:

  • active AND (salary >= 50000 OR (age NOT BETWEEN 20 AND 30))

  • age IN (20, 30, 40) AND salary BETWEEN (50000, 80000)

Criteria API

If SQL is not enough or programmable queries are preferred then JPA criteria like API can be used. Here is an example:

import com.hazelcast.core.IMap;
import com.hazelcast.query.Predicate;
import com.hazelcast.query.PredicateBuilder;
import com.hazelcast.query.EntryObject;

IMap map = Hazelcast.getMap("employee");

EntryObject e = new PredicateBuilder().getEntryObject();
Predicate predicate = e.is("active").and(e.get("age").lessThan(30));

Set<Employee> employees = (Set<Employee>) map.values(predicate);

Indexing

Hazelcast distributed queries will run on each member in parallel and only results will return the caller. When a query runs on a member, Hazelcast will iterate through the entire owned entries and find the matching ones. Can we make this even faster? Yes by indexing the mostly queried fields. Just like you would do for your database. Of course, indexing will add overhead for each write operation but queries will be a lot faster. If you are querying your map a lot then make sure to add indexes for most frequently queried fields. So if your active and age < 30 query, for example, is used a lot then make sure you add index for active and age fields. Here is how:

IMap imap = Hazelcast.getMap("employees");
imap.addIndex("age", true);        // ordered, since we have ranged queries for this field
imap.addIndex("active", false);    // not ordered, because boolean field cannot have range

API IMap.addIndex(fieldName, ordered) is used for adding index. For a each indexed field, if you have -ranged- queries such asage>30, age BETWEEN 40 AND 60 then ordered parameter should betrue, otherwise set it tofalse.

2.3.5. Near Cache

Map entries in Hazelcast are partitioned across the cluster. Imagine that you are reading key k so many times and k is owned by another member in your cluster. Each map.get(k) will be a remote operation; lots of network trips. If you have a map that is read-mostly then you should consider creating a Near Cache for the map so that reads can be much faster and consume less network traffic. All these benefits don't come free. When using near cache, you should consider the following issues:

  • JVM will have to hold extra cached data so it will increase the memory consumption.

  • If invalidation is turned on and entries are updated frequently, then invalidations will be costly.

  • Near cache breaks the strong consistency guarantees; you might be reading stale data.

Near cache is highly recommended for the maps that are read-mostly. Here is a near-cache configuration for a map :

<hazelcast>
    ...
    <map name="my-read-mostly-map">
        ...
        <near-cache>
            <!--
                Maximum size of the near cache. When max size is reached,
                cache is evicted based on the policy defined.
                Any integer between 0 and Integer.MAX_VALUE. 0 means
                Integer.MAX_VALUE. Default is 0.
            -->
            <max-size>5000</max-size>
            <!--
                Maximum number of seconds for each entry to stay in the near cache. Entries that are
                older than <time-to-live-seconds> will get automatically evicted from the near cache.
                Any integer between 0 and Integer.MAX_VALUE. 0 means infinite. Default is 0.
            -->
            <time-to-live-seconds>0</time-to-live-seconds>

            <!--
                Maximum number of seconds each entry can stay in the near cache as untouched (not-read).
                Entries that are not read (touched) more than <max-idle-seconds> value will get removed
                from the near cache.
                Any integer between 0 and Integer.MAX_VALUE. 0 means
                Integer.MAX_VALUE. Default is 0.
            -->
            <max-idle-seconds>60</max-idle-seconds>

            <!--
                Valid values are:
                NONE (no extra eviction, <time-to-live-seconds> may still apply),
                LRU  (Least Recently Used),
                LFU  (Least Frequently Used).
                NONE is the default.
                Regardless of the eviction policy used, <time-to-live-seconds> will still apply.
            -->
            <eviction-policy>LRU</eviction-policy>

            <!--
                Should the cached entries get evicted if the entries are changed (updated or removed).
                true of false. Default is true.
            -->
            <invalidate-on-change>true</invalidate-on-change>

        </near-cache>
    </map>
</hazelcast>

2.3.6. Entry Statistics

Hazelcast keeps extra information about each map entry such as creationTime, lastUpdateTime, lastAccessTime, number of hits, version, and this information is exposed to the developer via IMap.getMapEntry(key) call. Here is an example:

import com.hazelcast.core.Hazelcast;
import com.hazelcast.core.MapEntry;

MapEntry entry = Hazelcast.getMap("quotes").getMapEntry("1");
System.out.println ("size in memory  : " + entry.getCost();
System.out.println ("creationTime    : " + entry.getCreationTime();
System.out.println ("expirationTime  : " + entry.getExpirationTime();
System.out.println ("number of hits  : " + entry.getHits();
System.out.println ("lastAccessedTime: " + entry.getLastAccessTime();
System.out.println ("lastUpdateTime  : " + entry.getLastUpdateTime();
System.out.println ("version         : " + entry.getVersion();
System.out.println ("isValid         : " + entry.isValid();
System.out.println ("key             : " + entry.getKey();
System.out.println ("value           : " + entry.getValue();
System.out.println ("oldValue        : " + entry.setValue(newValue);

2.3.7. Indexing

Map entries can be indexed to be able to query faster. These indexes can be created using IMap API. But this usage has a limitation; all indexes must be created before any value is put into map. Sometimes by design adding an index to map may be impossible before any value is added. For example if a map has MapLoader that loads entries during map creation, then adding indexes to map becomes meaningless. To solve this problem, Hazelcast introduces defining IMap indexes in configuration.

  • Hazelcast XML configuration

    <map name="default">
        ...
        <indexes>
            <index ordered="false">name</index>
            <index ordered="true">age</index>
        </indexes>
    </map>

  • Config API

                            mapConfig.addMapIndexConfig(new MapIndexConfig("name", false));
                            mapConfig.addMapIndexConfig(new MapIndexConfig("age", true));
                        

  • Spring XML configuration

    <hz:map name="default">
        <hz:indexes>
            <hz:index attribute="name"/>
            <hz:index attribute="age" ordered="true"/>
        </hz:indexes>
    </hz:map>

To learn about wildcard configuration feature, see Wildcard Configuration page.

2.4. Distributed MultiMap

MultiMap is a specialized map where you can associate a key with multiple values. Just like any other distributed data structure implementation in Hazelcast, MultiMap is distributed/partitioned and thread-safe.

import com.hazelcast.core.MultiMap;
import com.hazelcast.core.Hazelcast;
import java.util.Collection;

// a multimap to hold <customerId, Order> pairs
MultiMap<String, Order> mmCustomerOrders = Hazelcast.getMultiMap("customerOrders");
mmCustomerOrders.put("1", new Order ("iPhone", 340));
mmCustomerOrders.put("1", new Order ("MacBook", 1200));
mmCustomerOrders.put("1", new Order ("iPod", 79));

// get orders of the customer with customerId 1.
Collection<Order> colOrders = mmCustomerOrders.get ("1");
for (Order order : colOrders) {
    // process order
}

// remove specific key/value pair
boolean removed = mmCustomerOrders.remove("1", new Order ("iPhone", 340));

2.5. Distributed Set

Distributed Set is distributed and concurrent implementation ofjava.util.Set. Set doesn't allow duplicate elements, so elements in the set should have proper hashCode and equals methods.

import com.hazelcast.core.Hazelcast;
import java.util.Set;
import java.util.Iterator;

java.util.Set set = Hazelcast.getSet("IBM-Quote-History");
set.add(new Price(10, time1));
set.add(new Price(11, time2));
set.add(new Price(12, time3));
set.add(new Price(11, time4));
//....
Iterator it = set.iterator();
while (it.hasNext()) { 
    Price price = (Price) it.next(); 
    //analyze
}

2.6. Distributed List

Distributed List is very similar to distributed set, but it allows duplicate elements.

import com.hazelcast.core.Hazelcast;
import java.util.List;
import java.util.Iterator;

java.util.List list = Hazelcast.getList("IBM-Quote-Frequency");
list.add(new Price(10));
list.add(new Price(11));
list.add(new Price(12));
list.add(new Price(11));
list.add(new Price(12));
        
//....
Iterator it = list.iterator();
while (it.hasNext()) { 
    Price price = (Price) it.next(); 
    //analyze
}

2.7. Distributed Lock

import com.hazelcast.core.Hazelcast;
import java.util.concurrent.locks.Lock;

Lock lock = Hazelcast.getLock(myLockedObject);
lock.lock();
try {
    // do something here
} finally {
    lock.unlock();
} 
 

java.util.concurrent.locks.Lock.tryLock() with timeout is also supported. All operations on the Lock that Hazelcast.getLock(Object obj) returns are cluster-wide and Lock behaves just like java.util.concurrent.lock.ReentrantLock.

if (lock.tryLock (5000, TimeUnit.MILLISECONDS)) {
    try {  
       // do some stuff here..  
    } 
    finally {  
      lock.unlock();  
    }   
} 

Locks are fail-safe. If a member holds a lock and some of the members go down, cluster will keep your locks safe and available. Moreover, when a member leaves the cluster, all the locks acquired by this dead member will be removed so that these locks can be available for live members immediately.

2.8. Distributed Events

Hazelcast allows you to register for entry events to get notified when entries added, updated or removed. Listeners are cluster-wide. When a member adds a listener, it is actually registering for events originated in any member in the cluster. When a new member joins, events originated at the new member will also be delivered. All events are ordered, meaning, listeners will receive and process the events in the order they are actually occurred.

import java.util.Queue;
import java.util.Map; 
import java.util.Set; 
import com.hazelcast.core.Hazelcast;
import com.hazelcast.core.ItemListener;
import com.hazelcast.core.EntryListener;
import com.hazelcast.core.EntryEvent; 

public class Sample implements ItemListener, EntryListener {

    public static void main(String[] args) { 
        Sample sample = new Sample();
        Queue queue = Hazelcast.getQueue ("default"); 
        Map   map   = Hazelcast.getMap   ("default"); 
        Set   set   = Hazelcast.getSet   ("default"); 
        //listen for all added/updated/removed entries
        queue.addItemListener(sample, true);
        set.addItemListener  (sample, true); 
        map.addEntryListener (sample, true);        
        //listen for an entry with specific key 
        map.addEntryListener (sample, "keyobj");        
    } 

    public void entryAdded(EntryEvent event) {
        System.out.println("Entry added key=" + event.getKey() + ", value=" + event.getValue());
    }

    public void entryRemoved(EntryEvent event) {
        System.out.println("Entry removed key=" + event.getKey() + ", value=" + event.getValue());
    }

    public void entryUpdated(EntryEvent event) {
        System.out.println("Entry update key=" + event.getKey() + ", value=" + event.getValue());
    } 

    public void entryEvicted(EntryEvent event) {
        System.out.println("Entry evicted key=" + event.getKey() + ", value=" + event.getValue());
    } 
    
    public void itemAdded(Object item) {
        System.out.println("Item added = " + item);
    }

    public void itemRemoved(Object item) {
        System.out.println("Item removed = " + item);
    }     
}
       

Chapter 3. Elastic Memory (Enterprise Edition Only)

By default, Hazelcast stores your distributed data (map entries, queue items) into Java heap which is subject to garbage collection. As your heap gets bigger, garbage collection might cause your application to pause tens of seconds, badly effecting your application performance and response times. Elastic Memory is Hazelcast with off-heap (direct) memory storage to avoid GC pauses. Even if you have terabytes of cache in-memory with lots of updates, GC will have almost no effect; resulting in more predictable latency and throughput.

Here are the steps to enable Elastic Memory:

  • Set the maximum direct memory JVM can allocate. Example java -XX:MaxDirectMemorySize=60G ...

  • Enable Elastic Memory by setting hazelcast.elastic.memory.enabled Hazelcast Config Property to true.

  • Set the total direct memory size for HazelcastInstance by setting hazelcast.elastic.memory.total.size Hazelcast Config Property. Size can be in MB or GB and abbreviation can be used, such as 60G and 500M.

  • Set the chunk size in KB by setting hazelcast.elastic.memory.chunk.size Hazelcast Config Property. Hazelcast will partition the entire offheap memory into chunks. Default chunk size is 1. Chunk size has to be power of 2 such as 1, 2, 4 and 8.

  • Configure maps you want them to use Elastic Memory by setting StorageType toOFFHEAP. Default value isHEAP.

    Using XML configuration:

    <hazelcast>
        ...
        <map name="default">
            ...
            <storage-type>OFFHEAP</storage-type>
        </map>
    </hazelcast>

    Using Config API:

    MapConfig mapConfig = new MapConfig();
    mapConfig.setStorageType(StorageType.OFFHEAP);

If NearCache is defined for a map, all near cached items are also going to be stored on the same off-heap. So off-heap storage is used for both near cache and distributed object storage.

Also see how to configure license key.

Chapter 4. Security (Enterprise Edition Only)

Hazelcast has an extensible, JAAS based security feature which can be used to authenticate both cluster members and clients and to do access control checks on client operations. Access control can be done according to endpoint principal and/or endpoint address. Security can be enabled and configured either in configuration xml or using Config api.

<hazelcast xsi:schemaLocation="http://www.hazelcast.com/schema/config
    http://www.hazelcast.com/schema/config/hazelcast-config-2.3.xsd"
    xmlns="http://www.hazelcast.com/schema/config"
    xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">
    
    ...
    <security enabled="true">
    ...
    </security>
</hazelcast>

Config cfg = new Config();
SecurityConfig securityCfg = cfg.getSecurityConfig();
securityCfg.setEnabled(true);

Also see how to configure license key.

4.1. Credentials

One of the key elements in Hazelcast security is Credentials object. It is used to carry all credentials of an endpoint (member or client). Credentials is an interface which extends Serializable and has three methods to be implemented. Users, according to their needs, can either implement Credentials interface or extend AbstractCredentials class which is an abstract implementation of Credentials.

package com.hazelcast.security;
...
public interface Credentials extends Serializable {

    String getEndpoint();

    void setEndpoint(String endpoint) ;
    
    String getPrincipal() ;    
}

Credentials.setEndpoint() method is called by Hazelcast when auth request arrives to node before authentication takes place.

package com.hazelcast.security;
...
public abstract class AbstractCredentials implements Credentials, DataSerializable {
    private transient String endpoint;
    private String principal;

    ...
}

UsernamePasswordCredentials, a custom implementation of Credentials can be found in Hazelcast com.hazelcast.security package. It is used by default configuration during authentication process of both members and clients.

package com.hazelcast.security;
...
public class UsernamePasswordCredentials extends Credentials {
    private byte[] password;
    ...
}

4.2. ClusterLoginModule

All security attributes are carried in Credentials object and Credentials is used by LoginModule s during authentication process. Accessing user supplied attributes from LoginModule s is done by CallbackHandler s. To provide access to Credentials object, Hazelcast uses its own specialized CallbackHandler. During initialization of LoginModules Hazelcast will pass this special CallbackHandlerinto LoginModule.initialize() method.

LoginModule implementations should create an instance of com.hazelcast.security.CredentialsCallback and call handle(Callback[] callbacks) method of CallbackHandler during login process. CredentialsCallback.getCredentials() will return supplied Credentials object.

public class CustomLoginModule implements LoginModule {
    CallbackHandler callbackHandler;
    Subject subject;

    public final void initialize(Subject subject, CallbackHandler callbackHandler,
        Map<String, ?> sharedState, Map<String, ?> options) {
        this.subject = subject;
        this.callbackHandler = callbackHandler;
    }

    public final boolean login() throws LoginException {
        CredentialsCallback callback = new CredentialsCallback();
        try {
            callbackHandler.handle(new Callback[]{callback});
            credentials = cb.getCredentials();
        } catch (Exception e) {
            throw new LoginException(e.getMessage());
        }
        ...
    }
...
}

* To use default Hazelcast permission policy, an instance of com.hazelcast.security.ClusterPrincipal that holding Credentials object must be created and added to Subject.principals onLoginModule.commit().

public class MyCustomLoginModule implements LoginModule {
...
    public boolean commit() throws LoginException {
        ...
        final Principal principal = new ClusterPrincipal(credentials);
        subject.getPrincipals().add(principal);
        
        return true;
    }
    ...
}

Hazelcast also has an abstract implementation of LoginModule that does callback and cleanup operations and holds resulting Credentials instance. LoginModules extending ClusterLoginModule can accessCredentials, Subject, LoginModule instances and options and sharedState maps. Extending ClusterLoginModule is recommended instead of implementing all required stuff.

package com.hazelcast.security;
...
public abstract class ClusterLoginModule implements LoginModule {

    protected abstract boolean onLogin() throws LoginException;
    protected abstract boolean onCommit() throws LoginException;
    protected abstract boolean onAbort() throws LoginException;
    protected abstract boolean onLogout() throws LoginException;

}

4.3. Cluster Member Security

Hazelcast supports standard Java Security (JAAS) based authentication between cluster members. You should configure one or moreLoginModules and an instance of com.hazelcast.security.ICredentialsFactory. Although Hazelcast has default implementations using cluster group and group-password and UsernamePasswordCredentials on authentication, it is advised to implement these according to specific needs and environment.

<security enabled="true">
    <member-credentials-factory class-name="com.hazelcast.examples.MyCredentialsFactory">
        <properties>
            <property name="property1">value1</property>
            <property name="property2">value2</property>
        </properties>
    </member-credentials-factory>
    <member-login-modules>
        <login-module class-name="com.hazelcast.examples.MyRequiredLoginModule" usage="required">
            <properties>
                <property name="property3">value3</property>
            </properties>
        </login-module>
        <login-module class-name="com.hazelcast.examples.MySufficientLoginModule" usage="sufficient">
            <properties>
                <property name="property4">value4</property>
            </properties>
        </login-module>
        <login-module class-name="com.hazelcast.examples.MyOptionalLoginModule" usage="optional">
            <properties>
                <property name="property5">value5</property>
            </properties>
        </login-module>
    </member-login-modules>
    ...
</security>

You can define as many asLoginModules you wanted in configuration. Those are executed in given order. Usage attribute has 4 values; 'required', 'requisite', 'sufficient' and 'optional' as defined in javax.security.auth.login.AppConfigurationEntry.LoginModuleControlFlag.

package com.hazelcast.security;
/**
 * ICredentialsFactory is used to create Credentials objects to be used
 * during node authentication before connection accepted by master node.
 */
public interface ICredentialsFactory {

    void configure(GroupConfig groupConfig, Properties properties);

    Credentials newCredentials();

    void destroy();
}

Properties defined in configuration are passed to ICredentialsFactory.configure() method as java.util.Properties and to LoginModule.initialize() method asjava.util.Map.

4.4. Native Client Security

Hazelcast's Client security includes both authentication and authorization.

4.4.1. Authentication

Authentication mechanism just works the same as cluster member authentication. Implementation of client authentication requires a Credentials and one or more LoginModule(s). Client side does not have/need a factory object to create Credentials objects likeICredentialsFactory. Credentials must be created at client side and sent to connected node during connection process.

<security enabled="true">
    <client-login-modules>
        <login-module class-name="com.hazelcast.examples.MyRequiredClientLoginModule" usage="required">
            <properties>
                <property name="property3">value3</property>
            </properties>
        </login-module>
        <login-module class-name="com.hazelcast.examples.MySufficientClientLoginModule" usage="sufficient">
            <properties>
                <property name="property4">value4</property>
            </properties>
        </login-module>
        <login-module class-name="com.hazelcast.examples.MyOptionalClientLoginModule" usage="optional">
            <properties>
                <property name="property5">value5</property>
            </properties>
        </login-module>
    </client-login-modules>
    ...
</security>

You can define as many asLoginModules you wanted in configuration. Those are executed in given order. Usage attribute has 4 values; 'required', 'requisite', 'sufficient' and 'optional' as defined in javax.security.auth.login.AppConfigurationEntry.LoginModuleControlFlag.

final Credentials credentials = new UsernamePasswordCredentials("dev", "dev-pass");
HazelcastInstance client = HazelcastClient.newHazelcastClient(credentials, "localhost");

4.4.2. Authorization

Hazelcast client authorization is configured by a client permission policy. Hazelcast has a default permission policy implementation that uses permission configurations defined in Hazelcast security configuration. Default policy permission checks are done against instance types (map, queue...), instance names (map, queue etc. name), instance actions (put, get, remove, add...), client endpoint addresses and client principal defined by Credentials object. Instance and principal names and endpoint addresses can be defined as wildcards(*). Take a look at Wildcard Name Configuration and Newtwork Configuration pages.

<security enabled="true">
    <client-permissions>
        <!-- Principal 'admin' from endpoint '127.0.0.1' has all permissions. -->
        <all-permissions principal="admin">
            <endpoints>
                <endpoint>127.0.0.1</endpoint>
            </endpoints>
        </all-permissions>
        
        <!-- Principals named 'dev' from all endpoints have 'create', 'destroy', 
            'put', 'get' permissions for map named 'default'. -->
        <map-permission name="default" principal="dev">
            <actions>
                <action>create</action>
                <action>destroy</action>
                <action>put</action>
                <action>get</action>
            </actions>
        </map-permission>
        
        <!-- All principals from endpoints '127.0.0.1' or matching to '10.10.*.*' 
            have 'put', 'get', 'remove' permissions for map 
            whose name matches to 'com.foo.entity.*'. -->
        <map-permission name="com.foo.entity.*">
            <endpoints>
                <endpoint>10.10.*.*</endpoint>
                <endpoint>127.0.0.1</endpoint>
            </endpoints>
            <actions>
                <action>put</action>
                <action>get</action>
                <action>remove</action>
            </actions>
        </map-permission>
        
        <!-- Principals named 'dev' from endpoints matching to either 
            '192.168.1.1-100' or '192.168.2.*' 
            have 'create', 'offer', 'poll' permissions for all queues. -->
        <queue-permission name="*" principal="dev">
            <endpoints>
                <endpoint>192.168.1.1-100</endpoint>
                <endpoint>192.168.2.*</endpoint>
            </endpoints>
            <actions>
                <action>create</action>
                <action>offer</action>
                <action>poll</action>
            </actions>
        </queue-permission>
        
        <!-- All principals from all endpoints have transaction permission.-->
        <transaction-permission />
    </client-permissions>
</security>

Users also can define their own policy by implementing com.hazelcast.security.IPermissionPolicy.

package com.hazelcast.security;
/**
 * IPermissionPolicy is used to determine any Subject's 
 * permissions to perform a security sensitive Hazelcast operation.
 *
 */
public interface IPermissionPolicy {
    void configure(SecurityConfig securityConfig, Properties properties);
    
    PermissionCollection getPermissions(Subject subject, Class<? extends Permission> type);
    
    void destroy();
}

Permission policy implementations can access client-permissions in configuration by using SecurityConfig.getClientPermissionConfigs() during configure(SecurityConfig securityConfig, Properties properties) method is called by Hazelcast. IPermissionPolicy.getPermissions(Subject subject, Class<? extends Permission> type) method is used to determine a client request has been granted permission to do a security-sensitive operation. Permission policy should return a PermissionCollection containing permissions of given type for givenSubject. Hazelcast access controller will call PermissionCollection.implies(Permission) on returning PermissionCollection and will decide if current Subject has permitted to access to requested resources or not.

4.4.3. Permissions

  1. All Permission

    <all-permissions principal="principal">
        <endpoints>
            ...
        </endpoints>
    </all-permissions>
    

  2. Map Permission

    <map-permission name="name" principal="principal">
        <endpoints>
            ...
        </endpoints>
        <actions>
            ...
        </actions>
    </map-permission>
    

    Actions: all, create, destroy, put, get, remove, listen, lock, stats

  3. Queue Permission

    <queue-permission name="name" principal="principal">
        <endpoints>
            ...
        </endpoints>
        <actions>
            ...
        </actions>
    </queue-permission>
    

    Actions: all, create, destroy, offer, poll, get, remove, listen, stats

  4. Multimap Permission

    <multimap-permission name="name" principal="principal">
        <endpoints>
            ...
        </endpoints>
        <actions>
            ...
        </actions>
    </multimap-permission>
    

    Actions: all, create, destroy, put, get, remove, listen, lock, stats

  5. Topic Permission

    <topic-permission name="name" principal="principal">
        <endpoints>
            ...
        </endpoints>
        <actions>
            ...
        </actions>
    </topic-permission>
    

    Actions: create, destroy, publish, listen, stats

  6. List Permission

    <list-permission name="name" principal="principal">
        <endpoints>
            ...
        </endpoints>
        <actions>
            ...
        </actions>
    </list-permission>
    

    Actions: all, create, destroy, add, set, get, remove, listen

  7. Set Permission

    <set-permission name="name" principal="principal">
        <endpoints>
            ...
        </endpoints>
        <actions>
            ...
        </actions>
    </set-permission>
    

    Actions: all, create, destroy, add, get, remove, listen

  8. Lock Permission

    <lock-permission name="name" principal="principal">
        <endpoints>
            ...
        </endpoints>
        <actions>
            ...
        </actions>
    </lock-permission>
    

    Actions: all, create, destroy, lock, stats

  9. AtomicNumber Permission

    <atomic-number-permission name="name" principal="principal">
        <endpoints>
            ...
        </endpoints>
        <actions>
            ...
        </actions>
    </atomic-number-permission>
    

    Actions: all, create, destroy, increment, decrement, get, set, add, stats

  10. CountDownLatch Permission

    <countdown-latch-permission name="name" principal="principal">
        <endpoints>
            ...
        </endpoints>
        <actions>
            ...
        </actions>
    </countdown-latch-permission>
    

    Actions: all, create, destroy, countdown, set, stats

  11. Semaphore Permission

    <semaphore-permission name="name" principal="principal">
        <endpoints>
            ...
        </endpoints>
        <actions>
            ...
        </actions>
    </semaphore-permission>
    

    Actions: all, create, destroy, acquire, release, drain, stats

  12. Executor Service Permission

    <executor-service-permission name="name" principal="principal">
        <endpoints>
            ...
        </endpoints>
        <actions>
            ...
        </actions>
    </executor-service-permission>
    

    Actions: all, create, destroy, execute

  13. Listener Permission

    <listener-permission name="name" principal="principal">
        <endpoints>
            ...
        </endpoints>
    </listener-permission>
    

    Names: all, instance, member

  14. Transaction Permission

    <transaction-permission principal="principal">
        <endpoints>
            ...
        </endpoints>
    </transaction-permission>
    

Chapter 5. Data Affinity

Co-location of related data and computation!

Hazelcast has a standard way of finding out which member owns/manages each key object. Following operations will be routed to the same member, since all of them are operating based on the same key, "key1".

        
Hazelcast.getMap("mapa").put("key1", value);
Hazelcast.getMap("mapb").get("key1");
Hazelcast.getMap("mapc").remove("key1");
// since map names are different, operation will be manipulating
// different entries, but the operation will take place on the
// same member since the keys ("key1") are the same

Hazelcast.getLock ("key1").lock();
// lock operation will still execute on the same member of the cluster
// since the key ("key1") is same

Hazelcast.getExecutorService().execute(new DistributedTask(runnable, "key1"));
// distributed execution will execute the 'runnable' on the same member
// since "key1" is passed as the key.
    

So when the keys are the same then entries are stored on the same node. But we sometimes want to have related entries stored on the same node. Consider customer and his/her order entries. We would have customers map with customerId as the key and orders map with orderId as the key. Since customerId and orderIds are different keys, customer and his/her orders may fall into different members/nodes in your cluster. So how can we have them stored on the same node? The trick here is to create an affinity between customer and orders. If we can somehow make them part of the same partition then these entries will be co-located. We achieve this by making orderIds PartitionAware

public class OrderKey implements Serializable, PartitionAware {
    int customerId;
    int orderId;

    public OrderKey(int orderId, int customerId) {
        this.customerId = customerId;
        this.orderId = orderId;
    }

    public int getCustomerId() {
        return customerId;
    }

    public int getOrderId() {
        return orderId;
    }

    public Object getPartitionKey() {
        return customerId;
    }

    @Override
    public String toString() {
        return "OrderKey{" +
                "customerId=" + customerId +
                ", orderId=" + orderId +
                '}';
    }
}
        

Notice that OrderKey implements PartitionAware and getPartitionKey() returns the customerId. This will make sure that Customer entry and its Orders are going to be stored on the same node.

Map mapCustomers = Hazelcast.getMap("customers")
Map mapOrders = Hazelcast.getMap("orders")
// create the customer entry with customer id = 1
mapCustomers.put(1, customer);
// now create the orders for this customer
mapOrders.put(new OrderKey(21, 1), order);
mapOrders.put(new OrderKey(22, 1), order);
mapOrders.put(new OrderKey(23, 1), order);
    

Let say you have a customers map where customerId is the key and the customer object is the value. and customer object contains the customer's orders. and let say you want to remove one of the orders of a customer and return the number of remaining orders. Here is how you would normally do it:

        
public static int removeOrder(long customerId, long orderId) throws Exception {
    IMap<Long, Customer> mapCustomers = Hazelcast.getMap("customers");
    mapCustomers.lock (customerId);
    Customer customer = mapCustomers. get(customerId);
    customer.removeOrder (orderId);
    mapCustomers.put(customerId, customer);
    mapCustomers.unlock(customerId);
    return customer.getOrderCount();
}

There are couple of things we should consider:

  1. There are four distributed operations there.. lock, get, put, unlock.. Can we reduce the number of distributed operations?

  2. Customer object may not be that big but can we not have to pass that object through the wire? Notice that, we are actually passing customer object through the wire twice; get and put.

So instead, why not moving the computation over to the member (JVM) where your customer data actually is. Here is how you can do this with distributed executor service:

  1. Send a PartitionAware Callable task.

  2. Callable does the deletion of the order right there and returns with the remaining order count.

  3. Upon completion of the Callable task, return the result (remaining order count). Plus you do not have to wait until the the task complete; since distributed executions are asynchronous, you can do other things in the meantime.

Here is a sample code:

public static int removeOrder(long customerId, long orderId) throws Exception {
    ExecutorService es = Hazelcast.getExecutorService();
    OrderDeletionTask task = new OrderDeletionTask(customerId, orderId);
    Future future = es.submit(task);
    int remainingOrders = future.get();
    return remainingOrders;
}

public static class OrderDeletionTask implements Callable<Integer>, PartitionAware, Serializable {

   private long customerId;
   private long orderId;

   public OrderDeletionTask() {
   }
   public OrderDeletionTask(long customerId, long orderId) {
       super();
       this.customerId = customerId;
       this.orderId = orderId;
   }
   public Integer call () {
       IMap<Long, Customer> mapCustomers = Hazelcast.getMap("customers");
       mapCustomers.lock (customerId);
       Customer customer = mapCustomers. get(customerId);
       customer.removeOrder (orderId);
       mapCustomers.put(customerId, customer);
       mapCustomers.unlock(customerId);
       return customer.getOrderCount();
   }

   public Object getPartitionKey() {
       return customerId;
   }
}
        

Benefits of doing the same operation with distributed ExecutorService based on the key are:

  • Only one distributed execution (es.submit(task)), instead of four.

  • Less data is sent over the wire.

  • Since lock/update/unlock cycle is done locally (local to the customer data), lock duration for the Customer entry is much less so enabling higher concurrency.

Chapter 6. Monitoring with JMX

  • Add the following system properties to enable jmx agent

    • -Dcom.sun.management.jmxremote

    • -Dcom.sun.management.jmxremote.port=_portNo_ (to specify jmx port) optional

    • -Dcom.sun.management.jmxremote.authenticate=false (to disable jmx auth) optional

  • Enable Hazelcast property hazelcast.jmx

    • using Hazelcast configuration (api, xml, spring)

    • or set system property -Dhazelcast.jmx=true

  • Use jconsole, jvisualvm (with mbean plugin) or another jmx-compliant monitoring tool.

Following attributes can be monitored:

  • Cluster

    • config

    • group name

    • count of members and their addresses (host:port)

    • operations: restart, shutdown cluster

  • Member

    • inet address

    • port

    • lite member state

  • Statistics

    • count of instances

    • number of instances created, destroyed since startup

    • max instances created, destroyed per second

  • AtomicNumber

    • name

    • actual value

    • operations: add, set, compareAndSet, reset

  • List, Set

    • name

    • size

    • items (as strings)

    • operations: clear, reset statistics

  • Map

    • name

    • size

    • operations: clear

  • Queue

    • name

    • size

    • received and served items

    • operations: clear, reset statistics

  • Topic

    • name

    • number of messages dispatched since creation, in last second

    • max messages dispatched per second

Chapter 7. Cluster Utilities

7.1. Cluster Interface

Hazelcast allows you to register for membership events to get notified when members added or removed. You can also get the set of cluster members.

import com.hazelcast.core.*;

Cluster cluster = Hazelcast.getCluster();
cluster.addMembershipListener(new MembershipListener(){
    public void memberAdded(MembershipEvent membersipEvent) {
        System.out.println("MemberAdded " + membersipEvent);
    }

    public void memberRemoved(MembershipEvent membersipEvent) {
        System.out.println("MemberRemoved " + membersipEvent);
    }
});

Member localMember  = cluster.getLocalMember();
System.out.println ("my inetAddress= " + localMember.getInetAddress());

Set setMembers  = cluster.getMembers();
for (Member member : setMembers) {
    System.out.println ("isLocalMember " + member.localMember());
    System.out.println ("member.inetaddress " + member.getInetAddress());
    System.out.println ("member.port " + member.getPort());
}

7.2. Cluster-wide Id Generator

Hazelcast IdGenerator creates cluster-wide unique IDs. Generated IDs are long type primitive values between 0 and Long.MAX_VALUE . Id generation occurs almost at the speed of AtomicLong.incrementAndGet() . Generated IDs are unique during the life cycle of the cluster. If the entire cluster is restarted, IDs start from 0 again.

import com.hazelcast.core.IdGenerator;
import com.hazelcast.core.Hazelcast;

IdGenerator idGenerator = Hazelcast.getIdGenerator("customer-ids");
long id = idGenerator.newId();

7.3. LiteMember

LiteMembers are members with no storage. If -Dhazelcast.lite.member=true JVM parameter is set, then the JVM will join the cluster as a 'lite member' which will not be a 'data partition' (no data on that node) but will have super fast access to the cluster just like any regular member does.

Chapter 8. Transactions

8.1. Transaction Interface

Hazelcast can be used in transactional context. Basically start a transaction, work with queues, maps, sets and do other things then commit/rollback in one shot.

import java.util.Queue;
import java.util.Map;
import java.util.Set;
import com.hazelcast.core.Hazelcast;
import com.hazelcast.core.Transaction; 

Queue queue = Hazelcast.getQueue("myqueue");
Map map     = Hazelcast.getMap  ("mymap");
Set set     = Hazelcast.getSet  ("myset");

Transaction txn = Hazelcast.getTransaction();
txn.begin();
try {    
    Object obj = queue.poll();
    //process obj
    map.put ("1", "value1");
    set.add ("value");
    //do other things..
    txn.commit();
}catch (Throwable t)  {
    txn.rollback();
}

Isolation is always READ_COMMITTED . If you are in a transaction, you can read the data in your transaction and the data that is already committed and if not in a transaction, you can only read the committed data. Implementation is different for queue and map/set. For queue operations (offer,poll), offered and/or polled objects are copied to the next member in order to safely commit/rollback. For map/set, Hazelcast first acquires the locks for the write operations (put, remove) and holds the differences (what is added/removed/updated) locally for each transaction. When transaction is set to commit, Hazelcast will release the locks and apply the differences. When rolling back, Hazelcast will simply releases the locks and discard the differences. Transaction instance is attached to the current thread and each Hazelcast operation checks if the current thread holds a transaction, if so, operation will be transaction aware. When transaction is committed, rolled back or timed out, it will be detached from the thread holding it.

8.2. J2EE Integration

Hazelcast can be integrated into J2EE containers via Hazelcast Resource Adapter ( hazelcast-ra.rar ). After proper configuration, Hazelcast can participate in standard J2EE transactions.

<%@page import="javax.resource.ResourceException" %>
<%@page import="javax.transaction.*" %>
<%@page import="javax.naming.*" %>
<%@page import="javax.resource.cci.*" %> 
<%@page import="java.util.*" %> 
<%@page import="com.hazelcast.core.Hazelcast" %> 

<%
UserTransaction txn = null;
Connection conn = null;
Queue queue = Hazelcast.getQueue ("default");
Map map     = Hazelcast.getMap   ("default");
Set set     = Hazelcast.getSet   ("default");
List list   = Hazelcast.getList  ("default");

try { 
    Context context = new InitialContext(); 
    txn = (UserTransaction) context.lookup("java:comp/UserTransaction"); 
    txn.begin(); 
    
    ConnectionFactory cf = (ConnectionFactory) context.lookup ("java:comp/env/HazelcastCF"); 
    conn = cf.getConnection();  
    
    queue.offer("newitem");
    map.put ("1", "value1");
    set.add ("item1");
    list.add ("listitem1");
    
    txn.commit(); 
} catch (Throwable e) { 
    if (txn != null) {
        try {
            txn.rollback();
        } catch (Exception ix) {ix.printStackTrace();};
    }
    e.printStackTrace();
} finally { 
    if (conn != null) {
        try {
            conn.close();
        } catch (Exception ignored) {};
    }
}
%>

8.2.1. Resource Adapter Configuration

Deploying and configuring Hazelcast resource adapter is no different than any other resource adapter since it is a standard JCA resource adapter but resource adapter installation and configuration is container specific, so please consult your J2EE vendor documentation for details. Most common steps are:

  1. Add the hazelcast.jar to container's classpath. Usually there is a lib directory that is loaded automatically by the container on startup.

  2. Deployhazelcast-ra.rar. Usually there is a some kind of deploy directory. Name of the directory varies by container.

  3. Make container specific configurations when/after deployinghazelcast-ra.rar. Besides container specific configurations, JNDI name for Hazelcast resource is set.

  4. Configure your application to use the Hazelcast resource. Updating web.xml and/or ejb-jar.xml to let container know that your application will use the Hazelcast resource and define the resource reference.

  5. Make container specific application configuration to specify JNDI name used for the resource in the application.

8.2.2. Sample Glassfish v3 Web Application Configuration

  1. Place the hazelcast-<version>.jar into GLASSFISH_HOME/glassfish/domains/domain1/lib/ext/ directory.

  2. Place the hazelcast-ra-<version>.rar into GLASSFISH_HOME/glassfish/domains/domain1/autodeploy/ directory

  3. Add the following lines to the web.xml file.

    <resource-ref>
        <res-ref-name>HazelcastCF</res-ref-name>
        <res-type>com.hazelcast.jca.ConnectionFactoryImpl</res-type>
        <res-auth>Container</res-auth>
    </resource-ref>
    

Notice that we didn't have to put sun-ra.xml into the rar file because it comes with the hazelcast-ra-<version>.rar file already.

If Hazelcast resource is used from EJBs, you should configure ejb-jar.xml for resource reference and JNDI definitions, just like we did forweb.xml.

8.2.3. Sample JBoss Web Application Configuration

  • Place the hazelcast-<version>.jar into JBOSS_HOME/server/deploy/default/lib directory.

  • Place the hazelcast-ra-<version>.rar into JBOSS_HOME/server/deploy/default/deploy directory

  • Create a hazelcast-ds.xml at JBOSS_HOME/server/deploy/default/deploy directory containing the following content. Make sure to set the rar-name element tohazelcast-ra-<version>.rar.

    <?xml version="1.0" encoding="UTF-8"?>
    <!DOCTYPE connection-factories
      PUBLIC "-//JBoss//DTD JBOSS JCA Config 1.5//EN"
      "http://www.jboss.org/j2ee/dtd/jboss-ds_1_5.dtd">
    
    <connection-factories> 
     <tx-connection-factory>
          <local-transaction/>
          <track-connection-by-tx>true</track-connection-by-tx>
          <jndi-name>HazelcastCF</jndi-name>
          <rar-name>hazelcast-ra-<version>.rar</rar-name>
          <connection-definition>
               javax.resource.cci.ConnectionFactory
          </connection-definition> 
      </tx-connection-factory>
    </connection-factories>
    

  • Add the following lines to the web.xml file.

    <resource-ref>
        <res-ref-name>HazelcastCF</res-ref-name>
        <res-type>com.hazelcast.jca.ConnectionFactoryImpl</res-type>
        <res-auth>Container</res-auth>
    </resource-ref>
    

  • Add the following lines to the jboss-web.xml file.

    <resource-ref>
        <res-ref-name>HazelcastCF</res-ref-name>
        <jndi-name>java:HazelcastCF</jndi-name>
    </resource-ref> 
    

If Hazelcast resource is used from EJBs, you should configure ejb-jar.xml and jboss.xml for resource reference and JNDI definitions.

Chapter 9. Distributed Executor Service

One of the coolest new futures of Java 1.5 is the Executor framework, which allows you to asynchronously execute your tasks, logical units of works, such as database query, complex calculation, image rendering etc. So one nice way of executing such tasks would be running them asynchronously and doing other things meanwhile. When ready, get the result and move on. If execution of the task takes longer than expected, you may consider canceling the task execution. In Java Executor framework, tasks are implemented as java.util.concurrent.Callable andjava.util.Runnable.

import java.util.concurrent.Callable;
import java.io.Serializable;

public class Echo implements Callable<String>, Serializable {
    String input = null;

    public Echo() {
    }

    public Echo(String input) {
        this.input = input;
    }

    public String call() {
        return Hazelcast.getCluster().getLocalMember().toString() + ":" + input;
    }
}

Echo callable above, for instance, in its call() method, is returning the local member and the input passed in. Remember that Hazelcast.getCluster().getLocalMember() returns the local member and toString() returns the member's address (ip + port) in String form, just to see which member actually executed the code for our example. Of course, call() method can do and return anything you like. Executing a task by using executor framework is very straight forward. Simply obtain a ExecutorService instance, generally via Executors and submit the task which returns aFuture. After executing task, you don't have to wait for execution to complete, you can process other things and when ready use the future object to retrieve the result as show in code below.

ExecutorService executorService = Executors.newSingleThreadExecutor();
Future<String> future = executorService.submit (new Echo("myinput"));
//while it is executing, do some useful stuff
//when ready, get the result of your execution
String result = future.get();

9.1. Distributed Execution

Distributed executor service is a distributed implementation of java.util.concurrent.ExecutorService. It allows you to execute your code in cluster. In this chapter, all the code samples are based on the Echo class above. Please note that Echo class is Serializable . You can ask Hazelcast to execute your code (Runnable, Callable):

  • on a specific cluster member you choose.

  • on the member owning the key you choose.

  • on the member Hazelcast will pick.

  • on all or subset of the cluster members.

import com.hazelcast.core.Member;
import com.hazelcast.core.Hazelcast;
import com.hazelcast.core.MultiTask;
import com.hazelcast.core.DistributedTask;
import java.util.concurrent.ExecutorService;   
import java.util.concurrent.FutureTask;   
import java.util.concurrent.Future;   
import java.util.Set;

public void echoOnTheMember(String input, Member member) throws Exception {
   FutureTask<String> task = new DistributedTask<String>(new Echo(input), member);
   ExecutorService executorService = Hazelcast.getExecutorService();
   executorService.execute(task);
   String echoResult = task.get();
}

public void echoOnTheMemberOwningTheKey(String input, Object key) throws Exception {
   FutureTask<String> task = new DistributedTask<String>(new Echo(input), key);
   ExecutorService executorService = Hazelcast.getExecutorService();
   executorService.execute(task);
   String echoResult = task.get();
}

public void echoOnSomewhere(String input) throws Exception { 
   ExecutorService executorService = Hazelcast.getExecutorService();
   Future<String> task = executorService.submit(new Echo(input));
   String echoResult = task.get();
}

public void echoOnMembers(String input, Set<Member> members) throws Exception {
   MultiTask<String> task = new MultiTask<String>(new Echo(input), members);
   ExecutorService executorService = Hazelcast.getExecutorService();
   executorService.execute(task);
   Collection<String> results = task.get();
} 

Note that you can obtain the set of cluster members via Hazelcast.getCluster().getMembers() call. You can also extend the MultiTask class to override set(V result), setException(Throwable exception), done() methods for custom behaviour. Just like java.util.concurrent.FutureTask.get() , MultiTask.get() will throw java.util.concurrent.ExecutionException if any of the executions throws exception.

9.2. Execution Cancellation

What if the code you execute in cluster takes longer than acceptable. If you cannot stop/cancel that task it will keep eating your resources. Standard Java executor framework solves this problem with by introducing cancel() api and 'encouraging' us to code and design for cancellations, which is highly ignored part of software development.

public class Fibonacci<Long> implements Callable<Long>, Serializable {
    int input = 0; 

    public Fibonacci() { 
    } 

    public Fibonacci(int input) { 
        this.input = input;
    } 

    public Long call() {
        return calculate (input);
    }

    private long calculate (int n) {
        if (Thread.currentThread().isInterrupted()) return 0;
        if (n <= 1) return n;
        else return calculate(n-1) + calculate(n-2);
    }
}

The callable class above calculates the fibonacci number for a given number. In the calculate method, we are checking to see if the current thread is interrupted so that code can be responsive to cancellations once the execution started. Following fib() method submits the Fibonacci calculation task for number 'n' and waits maximum 3 seconds for result. If the execution doesn't complete in 3 seconds, future.get() will throw TimeoutException and upon catching it we interruptibly cancel the execution for saving some CPU cycles.

long fib(int n) throws Exception {
    ExecutorService es = Hazelcast.getExecutorService();
    Future future = es.submit(new Fibonacci(n));  
    try {
        return future.get(3, TimeUnit.SECONDS);
    } catch (TimeoutException e) {
        future.cancel(true);            
    }
    return -1;
}

fib(20) will probably will take less than 3 seconds but fib(50) will take way longer. (This is not the example for writing better fibonacci calculation code but for showing how to cancel a running execution that takes too long.) future.cancel(false) can only cancel execution before it is running (executing) but future.cancel(true) can interrupt running executions if your code is able to handle the interruption. So if you are willing to be able to cancel already running task then your task has to be designed to handle interruption. If calculate (int n) method didn't have if (Thread.currentThread().isInterrupted()) line, then you wouldn't be able to cancel the execution after it started.

9.3. Execution Callback

ExecutionCallback allows you to asynchronously get notified when the execution is done. When implementing ExecutionCallback.done(Future) method, you can check if the task is already cancelled.

import com.hazelcast.core.Hazelcast;
import com.hazelcast.core.ExecutionCallback;
import com.hazelcast.core.DistributedTask;
import java.util.concurrent.ExecutorService;    
import java.util.concurrent.Future; 
        
ExecutorService es = Hazelcast.getExecutorService();
DistributedTask<String> task = new DistributedTask<String>(new Fibonacci<Long>(10));
task.setExecutionCallback(new ExecutionCallback<Long> () {
    public void done (Future<Long> future) { 
        try {
            if (! future.isCancelled()) {
                System.out.println("Fibonacci calculation result = " + future.get());
            }
        } catch (Exception e) {
            e.printStackTrace();
        }
    }
});
es.execute(task);

You could have achieved the same results by extending DistributedTask and overriding the DistributedTask.done() method.

import com.hazelcast.core.Hazelcast;
import com.hazelcast.core.DistributedTask;
import java.util.concurrent.ExecutorService;    
import java.util.concurrent.Future; 
        
ExecutorService es = Hazelcast.getExecutorService();
es.execute(new DistributedTask<String>(new Fibonacci<Long>(10)) {
    public void done () { 
        try {
            if (! isCancelled()) {
                System.out.println("Fibonacci calculation result = " + get());
            }
        } catch (Exception e) {
            e.printStackTrace();
        }
    }
}); 

Chapter 10. Http Session Clustering with HazelcastWM

Say you have more than one web servers (A, B, C) with a load balancer in front of them. If server A goes down then your users on that server will be directed to one of the live servers (B or C) but their sessions will be lost! So we have to have all these sessions backed up somewhere if we don't want to lose the sessions upon server crashes. Hazelcast WM allows you to cluster user http sessions automatically. Followings are required for enabling Hazelcast Session Clustering:

  • Target application or web server should support Java 1.5+

  • Target application or web server should support Servlet 2.4+ spec

  • Session objects that needs to be clustered have to be Serializable

Here are the steps to setup Hazelcast Session Clustering:

  1. Put the hazelcast and hazelcast-wm jars in your WEB-INF/lib directory. Optionally if you wish to connect to a cluster as a client add hazelcast-client as well.

  2. Put the following xml into web.xml file. Make sure Hazelcast filter is placed before all the other filters if any; put it at the top for example.

    <filter>
        <filter-name>hazelcast-filter</filter-name>
        <filter-class>com.hazelcast.web.WebFilter</filter-class>
        <!--
            Name of the distributed map storing
            your web session objects
        -->
        <init-param>
            <param-name>map-name</param-name>
            <param-value>my-sessions</param-value>
        </init-param>
        <!--
            How is your load-balancer configured?
            stick-session means all requests of a session
            is routed to the node where the session is first created.
            This is excellent for performance.
            If sticky-session is set to false, when a session is updated
            on a node, entry for this session on all other nodes is invalidated.
            You have to know how your load-balancer is configured before
            setting this parameter. Default is true.
        -->
        <init-param>
            <param-name>sticky-session</param-name>
            <param-value>true</param-value>
        </init-param>
        <!--
            Name of session id cookie
        -->
        <init-param>
            <param-name>cookie-name</param-name>
            <param-value>hazelcast.sessionId</param-value>
        </init-param>
        <!--
            Are you debugging? Default is false.
        -->
        <init-param>
            <param-name>debug</param-name>
            <param-value>true</param-value>
        </init-param>
        <!--
            Configuration xml location;
                * as servlet resource OR
                * as classpath resource OR
                * as URL
            Default is one of hazelcast-default.xml
            or hazelcast.xml in classpath.
        -->
        <init-param>
            <param-name>config-location</param-name>
            <param-value>/WEB-INF/hazelcast.xml</param-value>
        </init-param>
        <!--
            Do you want to use an existing HazelcastInstance?
            Default is null.
        -->
        <init-param>
            <param-name>instance-name</param-name>
            <param-value>default</param-value>
        </init-param>
    

        <!--
            Do you want to connect as a client to an existing cluster?
            Default is false.
        -->
        <init-param>
            <param-name>use-client</param-name>
            <param-value>false</param-value>
        </init-param>
        <!--
            Client configuration location;
                * as servlet resource OR
                * as classpath resource OR
                * as URL
            Default is null.
        -->
        <init-param>
            <param-name>client-config-location</param-name>
            <param-value>/WEB-INF/hazelcast-client.properties</param-value>
        </init-param>
        <!--
            Do you want to shutdown HazelcastInstance during
            web application undeploy process?
            Default is true.
        -->
        <init-param>
            <param-name>shutdown-on-destroy</param-name>
            <param-value>true</param-value>
        </init-param>
    </filter>
    <filter-mapping>
        <filter-name>hazelcast-filter</filter-name>
        <url-pattern>/*</url-pattern>
        <dispatcher>FORWARD</dispatcher>
        <dispatcher>INCLUDE</dispatcher>
        <dispatcher>REQUEST</dispatcher>
    </filter-mapping>
    
    <listener>
        <listener-class>com.hazelcast.web.SessionListener</listener-class>
    </listener>
    

  3. Package and deploy your war file as you would normally do.

It is that easy! All http requests will go through Hazelcast WebFilter and it will put the session objects into Hazelcast distributed map if needed.

Info about sticky-sessions:

Hazelcast holds whole session attributes in a distributed map and in local http session. Local session is required for fast access to data and distributed map is needed for fail-safety.

  • If sticky-session is not used, whenever a session a attribute is updated in a node (in both node local session and clustered cache), that attribute should be invalidated in all other nodes' local sessions, because now they have dirty value. So when a request arrives one of those other nodes that attribute value is fetched from clustered cache.

  • To overcome performance penalty of sending invalidation messages during updates, sticky-sessions can be used. If Hazelcast knows sessions are sticky, invalidation will not be send, because Hazelcast assumes there is no other local session at the moment. When a server is down, requests belonging to a session hold in that server will routed to other one and that server will fetch session data from clustered cache. That means using sticky-sessions, one will not suffer performance penalty of accessing clustered data and can benefit recover from a server failure.

Chapter 11. WAN Replication

There are cases where you would need to synchronize multiple clusters. Synchronization of clusters is named as WAN (Wide Area Network) Replication because it is mainly used for replicating different clusters running on WAN. Imagine having different clusters in New York, London and Tokyo. Each cluster would be operating at very high speed in their LAN (Local Area Network) settings but you would want some or all parts of the data in these clusters replicating to each other. So updates in Tokyo cluster goes to London and NY, in the meantime updates in New York cluster is synchronized to Tokyo and London.

You can setup active-passive WAN Replication where only one active node replicating its updates on the passive one. You can also setup active-active replication where each cluster is actively updating and replication to the other cluster(s).

In the active-active replication setup, there might be cases where each node is updating the same entry in the same named distributed map. Thus, conflicts will occur when merging. For those cases, conflict-resolution will be needed. Here is how you can setup WAN Replication for London cluster for instance:

<hazelcast>
    <wan-replication name="my-wan-cluster">
        <target-cluster group-name="tokyo" group-password="tokyo-pass">
            <replication-impl>com.hazelcast.impl.wan.WanNoDelayReplication</replication-impl>
            <end-points>
                <address>10.2.1.1:5701</address>
                <address>10.2.1.2:5701</address>
            </end-points>
        </target-cluster>
        <target-cluster group-name="london" group-password="london-pass">
            <replication-impl>com.hazelcast.impl.wan.WanNoDelayReplication</replication-impl>
            <end-points>
                <address>10.3.5.1:5701</address>
                <address>10.3.5.2:5701</address>
            </end-points>
        </target-cluster>
    </wan-replication>

    <network>
    ...
    </network>

</network>
...
</hazelcast>

This can be the configuration of the cluster running in NY, replicating to Tokyo and London. Tokyo and London clusters should have similar configurations if they are also active replicas.

If NY and London cluster configurations contain wan-replication element and Tokyo cluster doesn't then it means NY and London are active endpoints and Tokyo is passive endpoint.

As noted earlier you can have Hazelcast replicate some or all of the data in your clusters. You might have 5 different distributed maps but you might want only one of these maps replicating across clusters. So you mark which maps to replicate by adding wan-replication-ref element into map configuration.

<hazelcast>
    <wan-replication name="my-wan-cluster">
        ...
    </wan-replication>

    <network>
    ...
    </network>
    <map name="my-shared-map">
        ...
        <wan-replication-ref name="my-wan-cluster">
            <merge-policy>hz.PASS_THROUGH</merge-policy>
        </wan-replication-ref>
    </map>
</network>
...
</hazelcast>

Here we have my-shared-map is configured to replicate itself to the cluster targets defined in the wan-replication element.

Note that you will also need to define a merge policy for merging replica entries and resolving conflicts during the merge. Default merge policy is hz.PASS_THROUGH which will apply all in-coming updates as is.

Chapter 12. Configuration

Hazelcast can be configured through xml or using configuration api or even mix of both.

  1. Xml Configuration

    If you are using default Hazelcast instance (Hazelcast.getDefaultInstance()) or creating new Hazelcast instance with passing null parameter (Hazelcast.newHazelcastInstance(null)), Hazelcast will look into two places for the configuration file:

    • System property: Hazelcast will first check if "hazelcast.config" system property is set to a file path. Example: -Dhazelcast.config=C:/myhazelcast.xml.

    • Classpath: If config file is not set as a system property, Hazelcast will check classpath for hazelcast.xml file.

    If Hazelcast doesn't find any config file, it will happily start with default configuration (hazelcast-default.xml) located in hazelcast.jar. (Before configuring Hazelcast, please try to work with default configuration to see if it works for you. Default should be just fine for most of the users. If not, then consider custom configuration for your environment.)

    If you want to specify your own configuration file to create Config, Hazelcast supports several ways including filesystem, classpath, InputStream, URL etc.:

    • Config cfg = new XmlConfigBuilder(xmlFileName).build();

    • Config cfg = new XmlConfigBuilder(inputStream).build();

    • Config cfg = new ClasspathXmlConfig(xmlFileName);

    • Config cfg = new FileSystemXmlConfig(configFilename);

    • Config cfg = new UrlXmlConfig(url);

    • Config cfg = new InMemoryXmlConfig(xml);

  2. Programmatic Configuration

    To configure Hazelcast programatically, just instantiate a Config object and set/change its properties/attributes due to your needs.

    Config cfg = new Config();
    cfg.setPort(5900);
    cfg.setPortAutoIncrement(false);
            
    NetworkConfig network = cfg.getNetworkConfig();
    Join join = network.getJoin();
    join.getMulticastConfig().setEnabled(false);
    join.getTcpIpConfig().addMember("10.45.67.32").addMember("10.45.67.100")
                .setRequiredMember("192.168.10.100").setEnabled(true);
    network.getInterfaces().setEnabled(true).addInterface("10.45.67.*");
            
    MapConfig mapCfg = new MapConfig();
    mapCfg.setName("testMap");
    mapCfg.setBackupCount(2);
    mapCfg.getMaxSizeConfig().setSize(10000);
    mapCfg.setTimeToLiveSeconds(300);
            
    MapStoreConfig mapStoreCfg = new MapStoreConfig();
    mapStoreCfg.setClassName("com.hazelcast.examples.DummyStore").setEnabled(true);
    mapCfg.setMapStoreConfig(mapStoreCfg);
    
    NearCacheConfig nearCacheConfig = new NearCacheConfig();
    nearCacheConfig.setMaxSize(1000).setMaxIdleSeconds(120).setTimeToLiveSeconds(300);
    mapCfg.setNearCacheConfig(nearCacheConfig);
    
    cfg.addMapConfig(mapCfg);

After creating Config object, you can use it to initialize default Hazelcast instance or create a new Hazelcast instance.

  • Hazelcast.init(cfg);

  • Hazelcast.newHazelcastInstance(cfg);

HazelcastInstance with a name:

  • To create a named HazelcastInstance you should set instanceName of Config object.

    Config cfg = new Config();
    config.setInstanceName('my-instance');
    Hazelcast.newHazelcastInstance(config);

  • To retrieve an existing HazelcastInstance using its name, use;

    Hazelcast.getHazelcastInstanceByName('my-instance');

  • To retrieve all existingHazelcastInstances, use;

    Hazelcast.getAllHazelcastInstances();

12.1. Creating Separate Clusters

By specifying group-name and group-password, you can separate your clusters in a simple way; dev group, production group, test group, app-a group etc...

<hazelcast>
    <group>
        <name>dev</name>
        <password>dev-pass</password>
    </group>
    ...
</hazelcast>

You can also set the groupName with Config API. JVM can host multiple Hazelcast instances (nodes). Each node can only participate in one group and it only joins to its own group, does not mess with others. Following code creates 3 separate Hazelcast nodes, h1 belongs to app1 cluster, while h2 and h3 are belong to app2 cluster.

Config configApp1 = new Config();
configApp1.getGroupConfig().setName("app1");

Config configApp2 = new Config();
configApp2.getGroupConfig().setName("app2");

HazelcastInstance h1 = Hazelcast.newHazelcastInstance(configApp1);
HazelcastInstance h2 = Hazelcast.newHazelcastInstance(configApp2);
HazelcastInstance h3 = Hazelcast.newHazelcastInstance(configApp2);

12.2. Network Configuration

12.2.1. Configuring TCP/IP Cluster

If multicast is not preferred way of discovery for your environment, then you can configure Hazelcast for full TCP/IP cluster. As configuration below shows, while enable attribute of multicast is set to false, tcp-ip has to be set to true. For the none-multicast option, all or subset of cluster members' hostnames and/or ip addresses must be listed. Note that all of the cluster members don't have to be listed there but at least one of them has to be active in cluster when a new member joins. The tcp-ip tag accepts an attribute called "conn-timeout-seconds". The default value is 5. Increasing this value is recommended if you have many IP's listed and members can not properly build up the cluster.

<hazelcast>
    ...
    <network>
        <port auto-increment="true">5701</port>
        <join>
            <multicast enabled="false">
                <multicast-group>224.2.2.3</multicast-group>
                <multicast-port>54327</multicast-port>
            </multicast>
            <tcp-ip enabled="true">
                <hostname>machine1</hostname>
                <hostname>machine2</hostname>
                <hostname>machine3:5799</hostname>
                <interface>192.168.1.0-7</interface>     
                <interface>192.168.1.21</interface> 
            </tcp-ip>
        </join>
        ...
    </network>
    ...
</hazelcast>

12.2.2. Specifying Network Interfaces

You can also specify which network interfaces that Hazelcast should use. Servers mostly have more than one network interface so you may want to list the valid IPs. Range characters ('*' and '-') can be used for simplicity. So 10.3.10.*, for instance, refers to IPs between 10.3.10.0 and 10.3.10.255. Interface 10.3.10.4-18 refers to IPs between 10.3.10.4 and 10.3.10.18 (4 and 18 included). If network interface configuration is enabled (disabled by default) and if Hazelcast cannot find an matching interface, then it will print a message on console and won't start on that node.

<hazelcast>
    ...
    <network>
        ....
        <interfaces enabled="true">
            <interface>10.3.16.*</interface> 
            <interface>10.3.10.4-18</interface> 
            <interface>192.168.1.3</interface>         
        </interfaces>    
    </network>
    ...
</hazelcast> 

12.2.3. EC2 Auto Discovery

Hazelcast supports EC2 Auto Discovery as of 1.9.4. It is useful when you don't want or can't provide the list of possible IP addresses. Here is a sample configuration: Disable join over multicast and tcp/ip and enable aws. Also provide the credentials. The aws tag accepts an attribute called "conn-timeout-seconds". The default value is 5. Increasing this value is recommended if you have many IP's listed and members can not properly build up the cluster.

<join>
    <multicast enabled="false">
        <multicast-group>224.2.2.3</multicast-group>
        <multicast-port>54327</multicast-port>
    </multicast>
    <tcp-ip enabled="false">
        <interface>192.168.1.2</interface>
    </tcp-ip>
    <aws enabled="true">
        <access-key>my-access-key</access-key>
        <secret-key>my-secret-key</secret-key>
        <region>us-west-1</region>                              <!-- optional, default is us-east-1 -->
        <host-header>ec2.amazonaws.com</host-header>              <!-- optional, default is ec2.amazonaws.com.
                                                If set, region shouldn't be set as it will override this property -->
        <security-group-name>hazelcast-sg</security-group-name> <!-- optional -->
        <tag-key>type</tag-key>                                  <!-- optional -->
        <tag-value>hz-nodes</tag-value>                          <!-- optional -->
    </aws>
</join>

You need to add hazelcast-cloud.jar dependency into your project. Note that it is also bundled inside hazelcast-all.jar. hazelcast-cloud module doesn't depend on any other third party modules.

12.2.4. Network Partitioning (Split-Brain Syndrome)

Imagine that you have 10-node cluster and for some reason the network is divided into two in a way that 4 servers cannot see the other 6. As a result you ended up having two separate clusters; 4-node cluster and 6-node cluster. Members in each sub-cluster are thinking that the other nodes are dead even though they are not. This situation is called Network Partitioning (aka Split-Brain Syndrome).

Since it is a network failure, there is no way to avoid it programatically and your application will run as two separate independent clusters but we should be able answer the following questions: "What will happen after the network failure is fixed and connectivity is restored between these two clusters? Will these two clusters merge into one again? If they do, how are the data conflicts resolved, because you might end up having two different values for the same key in the same map?"

Here is how Hazelcast deals with it:

  1. The oldest member of the cluster checks if there is another cluster with the same group-name and group-password in the network.

  2. If the oldest member founds such cluster, then figures out which cluster should merge to the other.

  3. Each member of the merging cluster will do the followings

    • pause (HazelcastInstance.getLifecycleService().pause())

    • take locally owned map entries

    • close all its network connections (detach from its cluster)

    • join to the new cluster

    • send merge request for each its locally owned map entry

    • resume (HazelcastInstance.getLifecycleService().resume())

So each member of the merging cluster is actually rejoining to the new cluster and sending merge request for each its locally owned map entry.

Q: Which cluster will merge into the other?

A. Smaller cluster will merge into the bigger one. If they have equal number of members then a hashing algorithm determines the merging cluster.

Q. Each cluster may have different versions of the same key in the same map. How is the conflict resolved?

A. Destination cluster will decide how to handle merging entry based on the MergePolicy set for that map. There are built-in merge policies such as hz.NO_MERGE, hz.ADD_NEW_ENTRY and hz.LATEST_UPDATE but you can develop your own merge policy by implementingcom.hazelcast.merge.MergePolicy. You should register your custom merge policy in the configuration so that Hazelcast can find it by name.

public interface MergePolicy {
    /**
    * Returns the value of the entry after the merge
    * of entries with the same key. Returning value can be
    * You should consider the case where existingEntry is null.
    *
    * @param mapName       name of the map
    * @param mergingEntry  entry merging into the destination cluster
    * @param existingEntry existing entry in the destination cluster
    * @return final value of the entry. If returns null then no change on the entry.
    */
    Object merge(String mapName, MapEntry mergingEntry, MapEntry existingEntry);
}

Here is how merge policies are registered and specified per map.

<hazelcast>
    ...
    <map name="default">
        <backup-count>1</backup-count>
        <eviction-policy>NONE</eviction-policy>
        <max-size>0</max-size>
        <eviction-percentage>25</eviction-percentage>
        <!--
            While recovering from split-brain (network partitioning),
            map entries in the small cluster will merge into the bigger cluster
            based on the policy set here. When an entry merge into the
            cluster, there might an existing entry with the same key already.
            Values of these entries might be different for that same key.
            Which value should be set for the key? Conflict is resolved by
            the policy set here. Default policy is hz.ADD_NEW_ENTRY

            There are built-in merge policies such as
            hz.NO_MERGE      ; no entry will merge.
            hz.ADD_NEW_ENTRY ; entry will be added if the merging entry's key
                               doesn't exist in the cluster.
            hz.HIGHER_HITS   ; entry with the higher hits wins.
            hz.LATEST_UPDATE ; entry with the latest update wins.
        -->
        <merge-policy>MY_MERGE_POLICY</merge-policy>
    </map>

    <merge-policies>
        <map-merge-policy name="MY_MERGE_POLICY">
            <class-name>com.acme.MyOwnMergePolicy</class-name>
        </map-merge-policy>
    </merge-policies>
    ...
</hazelcast>

12.2.5. SSL

Hazelcast allows you to use SSL socket communication among all Hazelcast members. You need to implement com.hazelcast.nio.ssl.SSLContextFactory and configure SSL section in network configuration.

public class MySSLContextFactory implements SSLContextFactory {
    public void init(Properties properties) throws Exception {
    }

    public SSLContext getSSLContext() {
        ...
        SSLContext sslCtx = SSLContext.getInstance(protocol);
        return sslCtx;
    }
}

<hazelcast>
    ...
    <network>
        ...
        <ssl enabled="true">
            <factory-class-name>com.hazelcast.examples.MySSLContextFactory</factory-class-name>
            <properties>
                <property name="foo">bar</property>
            </properties>
        </ssl>
    </network>
    ...
</hazelcast>

Hazelcast provides a default SSLContextFactory; com.hazelcast.nio.ssl.BasicSSLContextFactory which uses configured keystore to initialize SSLContext. All required is to define keyStore and keyStorePassword. Also you can set keyManagerAlgorithm (default SunX509), trustManagerAlgorithm (default SunX509) and protocol (default TLS).

<hazelcast>
    ...
    <network>
        ...
        <ssl enabled="true">
            <factory-class-name>com.hazelcast.nio.ssl.BasicSSLContextFactory</factory-class-name>
            <properties>
                <property name="keyStore">keyStore</property>
                <property name="keyStorePassword">keyStorePassword</property>
                <property name="keyManagerAlgorithm">SunX509</property>
                <property name="trustManagerAlgorithm">SunX509</property>
                <property name="protocol">TLS</property>
            </properties>
        </ssl>
    </network>
    ...
</hazelcast>

You can also set keyStore and keyStorePassword through javax.net.ssl.keyStore and javax.net.ssl.keyStorePassword system properties. Note that, you can not use SSL when Hazelcast Encryption is enabled.

12.2.6. Encryption

Hazelcast allows you to encrypt entire socket level communication among all Hazelcast members. Encryption is based on Java Cryptography Architecture and both symmetric and asymmetric encryption are supported. In symmetric encryption, each node uses the same key, so the key is shared. Here is a sample configuration for symmetric encryption:

<hazelcast>
    ...
    <network>
        ...
        <!--
            Make sure to set enabled=true
            Make sure this configuration is exactly the same on
            all members
        -->
        <symmetric-encryption enabled="true">
            <!--
               encryption algorithm such as
               DES/ECB/PKCS5Padding,
               PBEWithMD5AndDES,
               Blowfish,
               DESede
            -->
            <algorithm>PBEWithMD5AndDES</algorithm>

            <!-- salt value to use when generating the secret key -->
            <salt>thesalt</salt>

            <!-- pass phrase to use when generating the secret key -->
            <password>thepass</password>

            <!-- iteration count to use when generating the secret key -->
            <iteration-count>19</iteration-count>
        </symmetric-encryption>
    </network>
    ...
</hazelcast>

In asymmetric encryption, public and private key pair is used. Data is encrypted with one of these keys and decrypted with the other. The idea is that each node has to have its own private key and other trusted members' public key. So that means, for each member, we should do the followings:

  • Pick a unique name for the member. We will use the name as the key alias. Let's name them as member1, member2...memberN.

  • Generate the keystore and the private key for the member1. keytool -genkey -alias member1 -keyalg RSA -keypass thekeypass -keystore keystore -storetype JKS Remember all the parameters you used here because you will need this information when you configure asymmetric-encryption in your hazelcast.xml file.

  • Create a public certificate file so that we can add it to the other members' keystore keytool -export -alias member1 -keypass thekeypass -storepass thestorepass -keystore keystore -rfc -file member1.cer

  • Now take all the other members' public certificates, and add (import) them into member1's keystore

     keytool -import -alias member2 -file member2.cer -keystore keystore -storepass thestorepass
    
     keytool -import -alias member3 -file member3.cer -keystore keystore -storepass thestorepass
    
     ...
    
     keytool -import -alias memberN -file memberN.cer -keystore keystore -storepass thestorepass
    
    

You should repeat these steps for each trusted member in your cluster. Here is a sample configuration for asymmetric encryption:

<hazelcast>
    ...
    <network>
        ...
        <!--
            Make sure to set enabled=true
        -->
        <asymmetric-encryption enabled="true">
            <!-- encryption algorithm -->
            <algorithm>RSA/NONE/PKCS1PADDING</algorithm>
            <!-- private key password -->
            <keyPassword>thekeypass</keyPassword>
            <!-- private key alias -->
            <keyAlias>member1</keyAlias>
            <!-- key store type -->
            <storeType>JKS</storeType>
            <!-- key store password -->
            <storePassword>thestorepass</storePassword>
            <!-- path to the key store --> 
            <storePath>keystore</storePath>
        </asymmetric-encryption>
    </network>
    ...
</hazelcast>

12.2.7. Socket Interceptor

Hazelcast allows you to intercept socket connections before a node joins to cluster or a client connects to a node. This provides ability to add custom hooks to join/connection procedure (like identity checking using Kerberos, etc.). You should implement com.hazelcast.nio.MemberSocketInterceptor for members and com.hazelcast.nio.SocketInterceptor for clients.

public class MySocketInterceptor implements MemberSocketInterceptor {
    public void init(SocketInterceptorConfig socketInterceptorConfig) {
        // initialize interceptor
    }

    void onConnect(Socket connectedSocket) throws IOException {
        // do something meaningful when connected
    }

    public void onAccept(Socket acceptedSocket) throws IOException {
        // do something meaningful when accepted a connection
    }
}

<hazelcast>
    ...
    <network>
        ...
       <socket-interceptor enabled="true">
           <class-name>com.hazelcast.examples.MySocketInterceptor</class-name>
           <properties>
                <property name="kerberos-host">kerb-host-name</property>
                <property name="kerberos-config-file">kerb.conf</property>
           </properties>
       </socket-interceptor>
    </network>
    ...
</hazelcast>


public class MyClientSocketInterceptor implements SocketInterceptor {
    void onConnect(Socket connectedSocket) throws IOException {
        // do something meaningful when connected
    }
}

ClientConfig clientConfig = new ClientConfig();
clientConfig.setGroupConfig(new GroupConfig("dev","dev-pass")).addAddress("10.10.3.4");

MyClientSocketInterceptor myClientSocketInterceptor = new MyClientSocketInterceptor();
clientConfig.setSocketInterceptor(myClientSocketInterceptor);
HazelcastInstance client = HazelcastClient.newHazelcastClient(clientConfig);

12.2.8. IPv6 Support

Hazelcast supports IPv6 addresses seamlessly. [IPv6 support has been switched off by default. See note below] All you need is to define IPv6 addresses or interfaces in network configuration. Only limitation at the moment is you can not define wildcard IPv6 addresses in TCP-IP join configuration. Interfaces section does not have this limitation, you can configure wildcard IPv6 interfaces same as IPv4 interfaces.

<hazelcast>
    ...
    <network>
        <port auto-increment="true">5701</port>
        <join>
            <multicast enabled="false">
                <multicast-group>FF02:0:0:0:0:0:0:1</multicast-group>
                <multicast-port>54327</multicast-port>
            </multicast>
            <tcp-ip enabled="true">
                <member>[fe80::223:6cff:fe93:7c7e]:5701</member>
                <interface>192.168.1.0-7</interface>
                <interface>192.168.1.*</interface>
                <interface>fe80:0:0:0:45c5:47ee:fe15:493a</interface>
            </tcp-ip>
        </join>
        <interfaces enabled="true">
            <interface>10.3.16.*</interface>
            <interface>10.3.10.4-18</interface>
            <interface>fe80:0:0:0:45c5:47ee:fe15:*</interface>
            <interface>fe80::223:6cff:fe93:0-5555</interface>
        </interfaces>
        ...
    </network>

JVM has two system properties for setting the preferred protocol stack —IPv4 or IPv6— as well as the preferred address family types —inet4 or inet6. On a dual stack machine IPv6 stack is preferred by default, this can be changed through java.net.preferIPv4Stack=<true|false> system property. And when querying name services JVM prefers IPv4 addressed over IPv6 addresses and will return an IPv4 address if possible. This can be changed through java.net.preferIPv6Addresses=<true|false> system property.

Also see additional details on IPv6 support in Java .

Note:

IPv6 support has been switched off by default, since some platforms have issues in use of IPv6 stack. And some other platforms such as Amazon AWS have no support at all. To enable IPv6 support, just set configuration property hazelcast.prefer.ipv4.stack to false. See Configuration Properties.

12.3. Partition Group Configuration

Hazelcast distributes key objects into partitions (blocks) using a consistent hashing algorithm and those partitions are assigned to nodes. That means an entry is stored in a node which is owner of partition to that entry's key is assigned. Number of total partitions is default 271 and can be changed with configuration property hazelcast.map.partition.count. Along with those partitions, there are also copies of them as backups. Backup partitions can have multiple copies due to backup count defined in configuration, such as first backup partition, second backup partition etc. As a rule, a node can not hold more than one copy of a partition (ownership or backup). By default Hazelcast distributes partitions and their backup copies randomly and equally among cluster nodes assuming all nodes in the cluster are identical.

What if some nodes share same JVM or physical machine or chassis and you want backups of these nodes to be assigned to nodes in another machine or chassis? What if processing or memory capacities of some nodes are different and you do not want equal number of partitions to be assigned to all nodes?

You can group nodes in same JVM (or physical machine) or nodes located in the same chassis. Or you can group nodes to create identical capacity. We call these groups as partition groups. This way partitions are assigned to those partition groups instead of single nodes. And backups of these partitions are located in another partition group.

When you enable partition grouping, Hazelcast presents two choices to configure partition groups at the moments.

  • First one is to group nodes automatically using IP addresses of nodes, so nodes sharing same network interface will be grouped together.

    <partition-group enabled="true" group-type="HOST_AWARE" />

    Config config = ...;
    PartitionGroupConfig partitionGroupConfig = config.getPartitionGroupConfig();
    partitionGroupConfig.setEnabled(true).setGroupType(MemberGroupType.HOST_AWARE);

  • Second one is custom grouping using Hazelcast's interface matching configuration. This way, you can add different and multiple interfaces to a group. You can also use wildcards in interface addresses.

    <partition-group enabled="true" group-type="CUSTOM">
        <member-group>
            <interface>10.10.0.*</interface>
            <interface>10.10.3.*</interface>
            <interface>10.10.5.*</interface>
        </member-group>
        <member-group>
            <interface>10.10.10.10-100</interface>
            <interface>10.10.1.*</interface>
            <interface>10.10.2.*</interface>
        </member-group
    </partition-group>

    Config config = ...;
    PartitionGroupConfig partitionGroupConfig = config.getPartitionGroupConfig();
    partitionGroupConfig.setEnabled(true).setGroupType(MemberGroupType.CUSTOM);
    
    MemberGroupConfig memberGroupConfig = new MemberGroupConfig();
    memberGroupConfig.addInterface("10.10.0.*")
        .addInterface("10.10.3.*").addInterface("10.10.5.*");
    
    MemberGroupConfig memberGroupConfig2 = new MemberGroupConfig();
    memberGroupConfig2.addInterface("10.10.10.10-100")
        .addInterface("10.10.1.*").addInterface("10.10.2.*");
    
    partitionGroupConfig.addMemberGroupConfig(memberGroupConfig);
    partitionGroupConfig.addMemberGroupConfig(memberGroupConfig2);

12.4. Listener Configurations

Hazelcast provides various event listener extensions to receive specific event types. These are

  • MembershipListener for cluster membership events

  • InstanceListener for distributed instance creation and destroy events

  • MigrationListener for partition migration start and complete events

  • LifecycleListener for HazelcastInstance lifecycle events

  • EntryListener for IMap and MultiMap entry events

  • ItemListener for IQueue, ISet and IList item events

  • MessageListener for ITopic message events

These listeners can be added to and removed from related object using Hazelcast API. Such as

MembershipListener listener = new MyMembershipListener();
hazelcastInstance.getCluster().addMembershipListener(listener);
hazelcastInstance.getCluster().removeMembershipListener(listener);

EntryListener listener = new MyEntryListener();
IMap map = hazelcastInstance.getMap("default");
map.addEntryListener(listener, true);
map.removeEntryListener(listener);

ItemListener listener = new MyItemListener();
IQueue queue = hazelcastInstance.getQueue("default");
queue.addItemListener(listener, true);
queue.removeItemListener(listener);

Downside of attaching listeners using API is possibility of missing events between creation of object and registering listener. To overcome this race condition Hazelcast introduces registration of listeners in configuration. Listeners can be registered using either Hazelcast XML configuration, Config API or Spring configuration.

  • MembershipListener

    Hazelcast XML configuration

    <listeners>
            <listener>com.hazelcast.examples.MembershipListener</listener>
    </listeners>

    Config API

    config.addListenerConfig(new ListenerConfig("com.hazelcast.examples.MembershipListener"));

    Spring XML configuration

    <hz:listeners>
        <hz:listener class-name="com.hazelcast.spring.DummyMembershipListener"/>
        <hz:listener implementation="dummyMembershipListener"/>
    </hz:listeners>

  • InstanceListener

    Hazelcast XML configuration

    <listeners>
            <listener>com.hazelcast.examples.InstanceListener</listener>
    </listeners>

    Config API

    config.addListenerConfig(new ListenerConfig("com.hazelcast.examples.InstanceListener"));

    Spring XML configuration

    <hz:listeners>
        <hz:listener class-name="com.hazelcast.spring.DummyInstanceListener"/>
        <hz:listener implementation="dummyInstanceListener"/>
    </hz:listeners>

  • MigrationListener

    Hazelcast XML configuration

    <listeners>
            <listener>com.hazelcast.examples.MigrationListener</listener>
    </listeners>

    Config API

    config.addListenerConfig(new ListenerConfig("com.hazelcast.examples.MigrationListener"));

    Spring XML configuration

    <hz:listeners>
        <hz:listener class-name="com.hazelcast.spring.DummyMigrationListener"/>
        <hz:listener implementation="dummyMigrationListener"/>
    </hz:listeners>

  • LifecycleListener

    Hazelcast XML configuration

    <listeners>
            <listener>com.hazelcast.examples.LifecycleListener</listener>
    </listeners>

    Config API

                            config.addListenerConfig(new ListenerConfig("com.hazelcast.examples.LifecycleListener"));

    Spring XML configuration

    <hz:listeners>
        <hz:listener class-name="com.hazelcast.spring.DummyLifecycleListener"/>
        <hz:listener implementation="dummyLifecycleListener"/>
    </hz:listeners>

  • EntryListener for IMap

    Hazelcast XML configuration

    <map name="default">
        ...
        <entry-listeners>
            <entry-listener include-value="true" local="false">com.hazelcast.examples.EntryListener</entry-listener>
        </entry-listeners>
    </map>

    Config API

    mapConfig.addEntryListenerConfig(new EntryListenerConfig("com.hazelcast.examples.EntryListener", false, false));

    Spring XML configuration

    <hz:map name="default">
        <hz:entry-listeners>
            <hz:entry-listener class-name="com.hazelcast.spring.DummyEntryListener" include-value="true"/>
            <hz:entry-listener implementation="dummyEntryListener" local="true"/>
        </hz:entry-listeners>
    </hz:map>

  • EntryListener for MultiMap

    Hazelcast XML configuration

    <multimap name="default">
        <value-collection-type>SET</value-collection-type>
        <entry-listeners>
            <entry-listener include-value="true" local="false">com.hazelcast.examples.EntryListener</entry-listener>
        </entry-listeners>
    </multimap>

    Config API

    multiMapConfig.addEntryListenerConfig(new EntryListenerConfig("com.hazelcast.examples.EntryListener", false, false));

    Spring XML configuration

    <hz:multimap name="default" value-collection-type="LIST">
        <hz:entry-listeners>
            <hz:entry-listener class-name="com.hazelcast.spring.DummyEntryListener" include-value="true"/>
            <hz:entry-listener implementation="dummyEntryListener" local="true"/>
        </hz:entry-listeners>
    </hz:multimap>

  • ItemListener for IQueue

    Hazelcast XML configuration

    <queue name="default">
        ...
        <item-listeners>
            <item-listener include-value="true">com.hazelcast.examples.ItemListener</item-listener>
        </item-listeners>
    </queue>

    Config API

    queueConfig.addItemListenerConfig(new ItemListenerConfig("com.hazelcast.examples.ItemListener", true));

    Spring XML configuration

    <hz:queue name="default" max-size-per-jvm="1000" backing-map-ref="default">
        <hz:item-listeners>
            <hz:item-listener class-name="com.hazelcast.spring.DummyItemListener" include-value="true"/>
        </hz:item-listeners>
    </hz:queue>

  • MessageListener for ITopic

    Hazelcast XML configuration

    <topic name="default">
        <message-listeners>
            <message-listener>com.hazelcast.examples.MessageListener</message-listener>
        </message-listeners>
    </topic>

    Config API

    topicConfig.addMessageListenerConfig(new ListenerConfig("com.hazelcast.examples.MessageListener"));

    Spring XML configuration

    <hz:topic name="default">
        <hz:message-listeners>
            <hz:message-listener class-name="com.hazelcast.spring.DummyMessageListener"/>
        </hz:message-listeners>
    </hz:topic>

12.5. Wildcard Configuration

Hazelcast supports wildcard configuration of Maps, Queues and Topics. Using an asterisk (*) character in the name, different instances of Maps, Queues and Topics can be configured by a single configuration.

Note that, with a limitation of a single usage, asterisk (*) can be placed anywhere inside the configuration name.

For instance a map named 'com.hazelcast.test.mymap' can be configured using one of these configurations;

<map name="com.hazelcast.test.*">
...
</map>

<map name="com.hazel*">
...
</map>

<map name="*.test.mymap">
...
</map>

<map name="com.*test.mymap">
...
</map>

Or a queue 'com.hazelcast.test.myqueue'

<queue name="*hazelcast.test.myqueue">
...
</queue>

<queue name="com.hazelcast.*.myqueue">
...
</queue>

12.6. Advanced Configuration Properties

There are some advanced configuration properties to tune some aspects of Hazelcast. These can be set as property name and value pairs through configuration xml, configuration API or JVM system property.

  • Configuration xml

    <hazelcast xsi:schemaLocation="http://www.hazelcast.com/schema/config http://www.hazelcast.com/schema/config/hazelcast-config-2.3.xsd"
        xmlns="http://www.hazelcast.com/schema/config"
        xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">
        ....
        <properties>
            <property name="hazelcast.property.foo">value</property>
                ....
        </properties>
    </hazelcast>

  • Configuration API

    Config cfg = new Config() ;
    cfg.setProperty("hazelcast.property.foo", "value");
                        

  • System Property

    1. Using JVM parameter: java -Dhazelcast.property.foo=value

    2. Using System class: System.setProperty("hazelcast.property.foo", "value");

Table 12.1. Properties Table
Property Name Description Value Type Default
hazelcast.memcache.enabled Enable Memcache client request listener service boolean true
hazelcast.rest.enabled Enable REST client request listener service boolean true
hazelcast.logging.type Name of logging framework type to send logging events. enum jdk
hazelcast.map.load.chunk.size Chunk size for MapLoader 's map initialization process (MapLoder.loadAllKeys()) integer 1000
hazelcast.in.thread.priority Hazelcast Input Thread priority integer 7
hazelcast.out.thread.priority Hazelcast Output Thread priority integer 7
hazelcast.service.thread.priority Hazelcast Service Thread priority integer 8
hazelcast.merge.first.run.delay.seconds Inital run delay of split brain/merge process in seconds integer 300
hazelcast.merge.next.run.delay.seconds Run interval of split brain/merge process in seconds integer 120
hazelcast.redo.wait.millis Wait time before a redo operation in milliseconds integer 500
hazelcast.redo.log.threshold Minimum number of redo(s) before logging. integer 10
hazelcast.redo.giveup.threshold Number of maximum redo calls before giving up and throwing OperationTimeoutException. integer 60
hazelcast.backup.redo.enabled Enable (strict) redo for backup operations. Backup operations may fail due to node failures, but generally during partition re-assignment process missing backups are restored. When 'hazelcast.backup.redo.enabled' is set to true, #hazelcast.partition.migration.interval should be 0. boolean false
hazelcast.max.operation.timeout Maximum operation timeout in milliseconds if no timeout is specified for an operation. (default 300 seconds) integer 300000
hazelcast.max.concurrent.operation.limit Max number of concurrent operations can be submitted to Hazelcast without throwing OperationRejectedException (negative means undefined). integer -1
hazelcast.socket.bind.any Bind node socket address to any local address boolean true
hazelcast.socket.receive.buffer.size Socket receive buffer size in KB integer 32
hazelcast.socket.send.buffer.size Socket send buffer size in KB integer 32
hazelcast.socket.keep.alive Socket set keep alive boolean true
hazelcast.socket.no.delay Socket set TCP no delay boolean true
hazelcast.prefer.ipv4.stack Prefer Ipv4 network interface when picking a local address. boolean true
hazelcast.shutdownhook.enabled Enable Hazelcast shutdownhook thread boolean true
hazelcast.wait.seconds.before.join Wait time before join operation integer 5
hazelcast.max.wait.seconds.before.join Maximum wait time before join operation integer 20
hazelcast.heartbeat.interval.seconds Heartbeat send interval in seconds integer 1
hazelcast.max.no.heartbeat.seconds Max timeout of heartbeat in seconds for a node to assume it is dead integer 300
hazelcast.icmp.enabled Enable ICMP ping boolean false
hazelcast.icmp.timeout ICMP timeout in ms int 1000
hazelcast.icmp.ttl ICMP TTL (maximum numbers of hops to try) int 0
hazelcast.prefer.ipv4.stack Prefer IPv4 Stack, don't use IPv6. See IPv6 doc. boolean true
hazelcast.initial.min.cluster.size Initial expected cluster size to wait before node to start completely integer 0
hazelcast.initial.wait.seconds Inital time in seconds to wait before node to start completely integer 0
hazelcast.restart.on.max.idle Restart node if service thread blocked for hazelcast.max.no.heartbeat.seconds boolean false
hazelcast.map.partition.count Distributed map partition count integer 271
hazelcast.map.max.backup.count Maximum map backup node count integer 5
hazelcast.map.remove.delay.seconds Remove delay time in seconds for dirty records integer 5
hazelcast.map.cleanup.delay.seconds Cleanup process delay time in seconds integer 10
hazelcast.executor.query.thread.count Query executor service max thread count integer 8
hazelcast.executor.event.thread.count Event executor service max thread count integer 16
hazelcast.executor.client.thread.count Client executor service max thread count integer 40
hazelcast.executor.store.thread.count Map store executor service max thread count integer 16
hazelcast.log.state Log cluster debug state periodically boolean false
hazelcast.jmx Enable JMX agent boolean false
hazelcast.jmx.detailed Enable detailed views on JMX boolean false
hazelcast.mc.map.excludes Comma seperated map names to exclude from Hazelcast Management Center CSV null
hazelcast.mc.queue.excludes Comma seperated queue names to exclude from Hazelcast Management Center CSV null
hazelcast.mc.topic.excludes Comma seperated topic names to exclude from Hazelcast Management Center CSV null
hazelcast.version.check.enabled Enable Hazelcast new version check on startup boolean true
hazelcast.topic.flow.control.enabled Enable waiting for the topic publish until messages are written through the sockets boolean true
hazelcast.mc.max.visible.instance.count Management Center maximum visible instance count integer 100
hazelcast.connection.monitor.interval Minimum interval to consider a connection error as critical in milliseconds. integer 100
hazelcast.connection.monitor.max.faults Maximum IO error count before disconnecting from a node. integer 3
hazelcast.partition.migration.interval Interval to run partition migration tasks in seconds. integer 0
hazelcast.partition.migration.timeout Timeout for partition migration tasks in seconds. integer 300
hazelcast.immediate.backup.interval Interval to run immediate backup tasks in seconds. integer 0
hazelcast.graceful.shutdown.max.wait Maximum wait seconds during graceful shutdown. integer 600
hazelcast.force.throw.interrupted.exception Force throw of RuntimeInterruptedException when a thread is interrupted, otherwise a warning log will be printed. boolean false
hazelcast.mc.url.change.enabled Management Center changing server url is enabled boolean true
hazelcast.elastic.memory.enabled Enable Hazelcast Elastic Memory off-heap storage boolean false
hazelcast.elastic.memory.total.size Hazelcast Elastic Memory storage total size in MB integer 128
hazelcast.elastic.memory.chunk.size Hazelcast Elastic Memory storage chunk size in KB integer 1
hazelcast.elastic.memory.shared.storage Enable Hazelcast Elastic Memory shared storage boolean false
hazelcast.enterprise.license.key Hazelcast Enterprise license key string null

12.7. Logging Configuration

Hazelcast has a flexible logging configuration and doesn't depend on any logging framework except JDK logging. It has in-built adaptors for a number of logging frameworks and also supports custom loggers by providing logging interfaces.

To use built-in adaptors you should set hazelcast.logging.type property to one of predefined types below.

  • jdk:JDK logging (default)

  • log4j:Log4j

  • slf4j:Slf4j

  • none:disable logging

You can set hazelcast.logging.type through configuration xml, configuration API or JVM system property.

  • Configuration xml

    <hazelcast xsi:schemaLocation="http://www.hazelcast.com/schema/config http://www.hazelcast.com/schema/config/hazelcast-config-2.3.xsd"
        xmlns="http://www.hazelcast.com/schema/config"
        xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">
    
        ....
    
        <properties>
            <property name="hazelcast.logging.type">jdk</property>
            ....
        </properties>
    </hazelcast>

  • Configuration API

    Config cfg = new Config() ;
    cfg.setProperty("hazelcast.logging.type", "log4j");
                        

  • System Property

    1. Using JVM parameter: java -Dhazelcast.logging.type=slf4j

    2. Using System class: System.setProperty("hazelcast.logging.type", "none");

To use custom logging feature you should implement com.hazelcast.logging.LoggerFactory and com.hazelcast.logging.ILogger interfaces and set system property hazelcast.logging.class to your custom LoggerFactory class name.

java -Dhazelcast.logging.class=foo.bar.MyLoggingFactory

You can also listen logging events generated by Hazelcast runtime by registeringLogListeners toLoggingService.

LogListener listener = new LogListener() {
    public void log(LogEvent logEvent) {
        // do something
    }
}
LoggingService loggingService = Hazelcast.getLoggingService();
loggingService.addLogListener(Level.INFO, listener):
        

Through the LoggingService you can get the current used ILogger implementation and log your own messages too.

12.8. Setting License Key (Enterprise Edition Only)

To be able to use Hazelcast Enterprise Edition, you need to set license key in configuration.

  • Hazelcast XML Configuration

    <hazelcast>
        ...
        <license-key>HAZELCAST_ENTERPRISE_LICENSE_KEY</license-key>
        ...
    </hazelcast>

  • Hazelcast Config API

    Config config = new Config();
    config.setLicenseKey("HAZELCAST_ENTERPRISE_LICENSE_KEY");
    

  • Spring XML Configuration

    <hz:config>
        ...
        <hz:license-key>HAZELCAST_ENTERPRISE_LICENSE_KEY</hz:license-key>
        ...
    </hazelcast>

  • JVM System Property

    -Dhazelcast.enterprise.license.key=HAZELCAST_ENTERPRISE_LICENSE_KEY

Chapter 13. Hibernate Second Level Cache

Hazelcast provides distributed second level cache for your Hibernate entities, collections and queries. Hazelcast has two implementations of Hibernate 2nd level cache, one for hibernate-pre-3.3 and one for hibernate-3.3.x versions. In your Hibernate configuration file (ex: hibernate.cfg.xml), add these properties;

  • To enable use of second level cache

    <property name="hibernate.cache.use_second_level_cache">true</property>
    

  • To enable use of query cache

    <property name="hibernate.cache.use_query_cache">true</property>
    

  • And to force minimal puts into cache

    <property name="hibernate.cache.use_minimal_puts">true</property>
    

  • To configure Hazelcast for Hibernate, it is enough to put configuration file named hazelcast.xml into root of your classpath. If Hazelcast can not find hazelcast.xml then it will use default configuration from hazelcast.jar.

  • You can define custom named Hazelcast configuration xml file with one of these Hibernate configuration properties.

    <property name="hibernate.cache.provider_configuration_file_resource_path">hazelcast-custom-config.xml</property>
    

    or

    <property name="hibernate.cache.hazelcast.configuration_file_path">hazelcast-custom-config.xml</property>
    

  • You can set up Hazelcast to connect cluster as LiteMember. LiteMember is a member of the cluster, it has socket connection to every member in the cluster and it knows where the data, but does not contain any data.

    <property name="hibernate.cache.hazelcast.use_lite_member">true</property>
    

  • You can set up Hazelcast to connect cluster as Native Client. Native client is not member and it connects to one of the cluster members and delegates all cluster wide operations to it. When the relied cluster member dies, client will transparently switch to another live member. (Native Client property takes precedence over LiteMember property.)

    <property name="hibernate.cache.hazelcast.use_native_client">true</property>
    

    To setup Native Client properly, you should add Hazelcastgroup-name, group-password and cluster member address properties. Native Client will connect to defined member and will get addresses of all members in the cluster. If the connected member will die or leave the cluster, client will automatically switch to another member in the cluster.

    <property name="hibernate.cache.hazelcast.native_client_address">10.34.22.15</property>
    <property name="hibernate.cache.hazelcast.native_client_group">dev</property>
    <property name="hibernate.cache.hazelcast.native_client_password">dev-pass</property>
    

    To use Native Client you should add hazelcast-client-<version>.jar into your classpath.

    Read more about NativeClient & LiteMember

  • If you are using one of Hibernate pre-3.3 version, add following property.

    <property name="hibernate.cache.provider_class">com.hazelcast.hibernate.provider.HazelcastCacheProvider</property>
    

  • If you are using Hibernate 3.3.x (or newer) version, you can choose to use either configuration property above (Hibernate has a built-in bridge to use old-style cache implementations) or following property.

    <property name="hibernate.cache.region.factory_class">com.hazelcast.hibernate.HazelcastCacheRegionFactory</property>
    

Hazelcast creates a seperate distributed map for each Hibernate cache region. So these regions can be configured easily via Hazelcast map configuration. You can definebackup,eviction, TTL and Near Cache properties.

Hibernate has four cache concurrency strategies:read-only,read-write, nonstrict-read-write andtransactional. But Hibernate does not forces cache providers to support all strategies. And Hazelcast supports first three (read-only,read-write,nonstrict-read-write) of these four strategies. Hazelcast has not support for transactional strategy yet.

  • If you are using xml based class configurations, you should add a cache element into your configuration with usage attribute with one ofread-only,read-write, nonstrict-read-write.

    <class name="eg.Immutable" mutable="false">
        <cache usage="read-only"/>
        .... 
    </class>
    
    <class name="eg.Cat" .... >
        <cache usage="read-write"/>
        ....
        <set name="kittens" ... >
            <cache usage="read-write"/>
            ....
        </set>
    </class>
    

  • If you are using Hibernate-Annotations then you can add class-cache or collection-cache element into your Hibernate configuration file with usage attribute with one ofread only,read/write,nonstrict read/write.

    <class-cache usage="read-only" class="eg.Immutable"/>
    <class-cache usage="read-write" class="eg.Cat"/>
    <collection-cache collection="eg.Cat.kittens" usage="read-write"/>
    

OR

  • Alternatively, you can put Hibernate Annotation's @Cache annotation on your entities and collections.

    @Cache(usage = CacheConcurrencyStrategy.READ_WRITE)
    public class Cat implements Serializable {
         ...
    }
    

And now last thing you should be aware of is to drop hazelcast-hibernate-<version>.jar into your classpath.

Additional Properties:

  • Accessing underlying HazelcastInstance

    Using com.hazelcast.hibernate.instance.HazelcastAccessor you can access underlying HazelcastInstance used by Hibernate SessionFactory.

    SessionFactory sessionFactory = ...;
    HazelcastInstance hazelcastInstance = HazelcastAccessor.getHazelcastInstance(sessionFactory);        
    

  • Changing/setting lock timeout value of read-write strategy

    Lock timeout value can be set using hibernate.cache.hazelcast.lock_timeout_in_seconds Hibernate property. Value should be in seconds and default value is 300 seconds.

  • Using named HazelcastInstance

    Instead of creating a new HazelcastInstance for eachSessionFactory, an existing instance can be used by setting hibernate.cache.hazelcast.instance_name Hibernate property toHazelcastInstance's name. For more information seeNamed HazelcastInstance.

  • Disabling shutdown during SessionFactory.close()

    Shutting down HazelcastInstance can be disabled during SessionFactory.close() by setting hibernate.cache.hazelcast.shutdown_on_session_factory_close Hibernate property to false. (In this case Hazelcast property hazelcast.shutdownhook.enabled should not be set to false.) Default value istrue.

Chapter 14. Spring Integration

14.1. Configuration

You can declare Hazelcast beans for Spring context using beans namespace (default spring beans namespace) as well to declare hazelcast maps, queues and others. Hazelcast-Spring integration requires either hazelcast-spring jar or hazelcast-all jar in the classpath.

<bean id="instance" class="com.hazelcast.core.Hazelcast" factory-method="newHazelcastInstance">
    <constructor-arg>
        <bean class="com.hazelcast.config.Config">
            <property name="groupConfig">
                <bean class="com.hazelcast.config.GroupConfig">
                    <property name="name" value="dev"/>
                    <property name="password" value="pwd"/>
                </bean>
            </property>
            <!-- and so on ... -->
        </bean>
    </constructor-arg>
</bean>

<bean id="map" factory-bean="instance" factory-method="getMap">
    <constructor-arg value="map"/>
</bean>

Hazelcast has Spring integration (requires version 2.5 or greater) since 1.9.1 using hazelcast namespace.

  • Add namespace xmlns:hz="http://www.hazelcast.com/schema/spring" to beans tag in context file:

    <beans xmlns="http://www.springframework.org/schema/beans"
           xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
           xmlns:hz="http://www.hazelcast.com/schema/spring"
           xsi:schemaLocation="http://www.springframework.org/schema/beans
                    http://www.springframework.org/schema/beans/spring-beans-2.5.xsd
                    http://www.hazelcast.com/schema/spring
                    http://www.hazelcast.com/schema/spring/hazelcast-spring-2.3.xsd">
    

  • Use hz namespace shortcuts to declare cluster, its items and so on.

After that you can configure Hazelcast instance (node):

<hz:hazelcast id="instance">
    <hz:config>
        <hz:group name="dev" password="password"/>
        <hz:network port="5701" port-auto-increment="false">
            <hz:join>
                <hz:multicast enabled="false"
                              multicast-group="224.2.2.3"
                              multicast-port="54327"/>
                <hz:tcp-ip enabled="true">
                    <hz:members>10.10.1.2, 10.10.1.3</hz:members>
                </hz:tcp-ip>
            </hz:join>
        </hz:network>
        <hz:map name="map"
                backup-count="2"
                max-size="0"
                eviction-percentage="30"
                read-backup-data="true"
                cache-value="true"
                eviction-policy="NONE"
                merge-policy="hz.ADD_NEW_ENTRY"/>
    </hz:config>
</hz:hazelcast>

You can easily configure map-store and near-cache too. (For map-store you should set either class-name or implementation attribute.)

<hz:config>
    <hz:map name="map1">
            <hz:near-cache time-to-live-seconds="0" max-idle-seconds="60"
               eviction-policy="LRU" max-size="5000"  invalidate-on-change="true"/>

            <hz:map-store enabled="true" class-name="com.foo.DummyStore"
                write-delay-seconds="0"/>
    </hz:map>

    <hz:map name="map2">
            <hz:map-store enabled="true" implementation="dummyMapStore"
                write-delay-seconds="0"/>
    </hz:map>

    <bean id="dummyMapStore" class="com.foo.DummyStore" />
</hz:config>

It's possible to use placeholders instead of concrete values. For instance, use property file app-default.properties for group configuration:

<bean class="org.springframework.beans.factory.config.PropertyPlaceholderConfigurer">
    <property name="locations">
        <list>
            <value>classpath:/app-default.properties</value>
        </list>
    </property>
</bean>

<hz:hazelcast id="instance">
    <hz:config>
        <hz:group
            name="${cluster.group.name}"
            password="${cluster.group.password}"/>
        <!-- ... -->
    </hz:config>
</hz:hazelcast>

Similar for client

<hz:client id="client"
    group-name="${cluster.group.name}" group-password="${cluster.group.password}">
    <hz:members>10.10.1.2:5701, 10.10.1.3:5701</hz:members>
</hz:client>

Hazelcast also supports lazy-init, scope and depends-on bean attributes.

<hz:hazelcast id="instance" lazy-init="true" scope="singleton">
    ...
</hz:hazelcast>

<hz:client id="client" scope="prototype" depends-on="instance">
    ...
</hz:client>

You can declare beans for the following Hazelcast objects:

  • map

  • multiMap

  • queue

  • topic

  • set

  • list

  • executorService

  • idGenerator

  • atomicNumber

  • semaphore

  • countDownLatch

  • lock

Example:

<hz:map id="map" instance-ref="client" name="map" lazy-init="true" />
<hz:multiMap id="multiMap" instance-ref="instance" name="multiMap" lazy-init="false" />
<hz:queue id="queue" instance-ref="client" name="queue" lazy-init="true" depends-on="instance"/>
<hz:topic id="topic" instance-ref="instance" name="topic" depends-on="instance, client"/>
<hz:set id="set" instance-ref="instance" name="set" />
<hz:list id="list" instance-ref="instance" name="list"/>
<hz:executorService id="executorService" instance-ref="client" name="executorService"/>
<hz:idGenerator id="idGenerator" instance-ref="instance" name="idGenerator"/>
<hz:atomicNumber id="atomicNumber" instance-ref="instance" name="atomicNumber"/>
<hz:atomicNumber id="semaphore" instance-ref="client" name="semaphore"/>
<hz:atomicNumber id="countDownLatch" instance-ref="instance" name="countDownLatch"/>
<hz:atomicNumber id="lock" instance-ref="client" name="lock"/>

Injecting Typed Collections/Maps

Spring tries to create a new Map/Collection instance and fill the new instance by iterating and converting values of the original Map/Collection (IMap, IQueue etc.) to required types when generic type parameters of the original Map/Collection and the target property/attribute do not match.

Since Hazelcast Maps/Collections are designed to hold very large data which a single machine can not carry, iterating through whole values can cause out of memory errors.

To avoid this issue either target property/attribute can be declared as un-typed Map/Collection

public class SomeBean {
    @Autowired
    IMap map; // instead of IMap<K, V> map

    @Autowired
    IQueue queue; // instead of IQueue<E> queue

    ...
}

or parameters of injection methods (constructor, setter) can be un-typed.

public class SomeBean {

    IMap<K, V> map;

    IQueue<E> queue;

    public SomeBean(IMap map) { // instead of IMap<K, V> map
        this.map = map;
    }

    ...

    public void setQueue(IQueue queue) { // instead of IQueue<E> queue
        this.queue = queue;
    }
    ...
}

For more info see Spring issue-3407 .

14.2. Spring Managed Context

It's often desired to access Spring managed beans, to apply bean properties or to apply factory callbacks such as ApplicationContextAware, BeanNameAware or to apply bean post-processing such as InitializingBean, @PostConstruct like annotations while using Hazelcast distributed ExecutorService or DistributedTasks or more generally any Hazelcast managed object. Achieving those features are as simple as adding @SpringAware annotation to your distributed object types. Once you have configured HazelcastInstance as explained in Spring configuration, just mark any distributed type with @SpringAware annotation.

<beans xmlns="http://www.springframework.org/schema/beans"
       xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
       xmlns:context="http://www.springframework.org/schema/context"
       xmlns:hz="http://www.hazelcast.com/schema/spring"
       xsi:schemaLocation="http://www.springframework.org/schema/beans
                http://www.springframework.org/schema/beans/spring-beans-3.0.xsd
		        http://www.springframework.org/schema/context
                http://www.springframework.org/schema/context/spring-context-3.0.xsd
                http://www.hazelcast.com/schema/spring
                http://www.hazelcast.com/schema/spring/hazelcast-spring-2.3.xsd">

    <context:annotation-config />

    <hz:hazelcast id="instance">
        <hz:config>
            <hz:group name="dev" password="password"/>
            <hz:network port="5701" port-auto-increment="false">
                <hz:join>
                    <hz:multicast enabled="false" />
                    <hz:tcp-ip enabled="true">
                        <hz:members>10.10.1.2, 10.10.1.3</hz:members>
                    </hz:tcp-ip>
                </hz:join>
            </hz:network>
            ...
        </hz:config>
    </hz:hazelcast>

    <bean id="someBean" class="com.hazelcast.examples.spring.SomeBean" scope="singleton" />
    ...
</beans>

ExecutorService example:

@SpringAware
public class SomeTask implements Callable<Long>, ApplicationContextAware, Serializable {

    private transient ApplicationContext context;

    private transient SomeBean someBean;

    public Long call() throws Exception {
        return someBean.value;
    }

    public void setApplicationContext(final ApplicationContext applicationContext)
        throws BeansException {
        context = applicationContext;
    }

    @Autowired
    public void setSomeBean(final SomeBean someBean) {
        this.someBean = someBean;
    }
}

HazelcastInstance hazelcast = (HazelcastInstance) context.getBean("hazelcast");
SomeBean bean = (SomeBean) context.getBean("someBean");

Future<Long> f = hazelcast.getExecutorService().submit(new SomeTask());
Assert.assertEquals(bean.value, f.get().longValue());

// choose a member
Member member = hazelcast.getCluster().getMembers().iterator().next();

Future<Long> f2 = (Future<Long>) hazelcast.getExecutorService()
    .submit(new DistributedTask<Long>(new SomeTask(), member));
Assert.assertEquals(bean.value, f2.get().longValue());

Distributed Map value example:

@SpringAware
@Component("someValue")
@Scope("prototype")
public class SomeValue implements Serializable, ApplicationContextAware {

    transient ApplicationContext context;

    transient SomeBean someBean;

    transient boolean init = false;

    public void setApplicationContext(final ApplicationContext applicationContext)
        throws BeansException {
        context = applicationContext;
    }

    @Autowired
    public void setSomeBean(final SomeBean someBean) {
        this.someBean = someBean;
    }

    @PostConstruct
    public void init() {
        someBean.doSomethingUseful();
        init = true;
    }
    ...
}

On Node-1;

HazelcastInstance hazelcast = (HazelcastInstance) context.getBean("hazelcast");
SomeValue value = (SomeValue) context.getBean("someValue")
IMap<String, SomeValue> map = hazelcast.getMap("values");
map.put("key", value);

On Node-2;

HazelcastInstance hazelcast = (HazelcastInstance) context.getBean("hazelcast");
IMap<String, SomeValue> map = hazelcast.getMap("values");
SomeValue value = map.get("key");
Assert.assertTrue(value.init);

Note that, Spring managed properties/fields are marked as transient.

14.3. Spring Cache

As of version 3.1, Spring Framework provides support for adding caching into an existing Spring application. To use Hazelcast as Spring cache provider, you should just define a com.hazelcast.spring.cache.HazelcastCacheManager bean and register it as Spring cache manager.

<cache:annotation-driven cache-manager="cacheManager" />

<hz:hazelcast id="hazelcast">
    ...
</hz:hazelcast>

<bean id="cacheManager" class="com.hazelcast.spring.cache.HazelcastCacheManager">
    <constructor-arg ref="instance"/>
</bean>

For more info see Spring Cache Abstraction .

14.4. Hibernate 2nd Level Cache Config

If you are using Hibernate with Hazelcast as 2nd level cache provider, you can easily create CacheProvider or RegionFactory instances within Spring configuration. That way it is possible to use same HazelcastInstance as Hibernate L2 cache instance.

<hz:hibernate-cache-provider id="cacheProvider" instance-ref="instance" />
...

<bean id="sessionFactory" class="org.springframework.orm.hibernate3.LocalSessionFactoryBean" scope="singleton">
    <property name="dataSource" ref="dataSource"/>
    <property name="cacheProvider" ref="cacheProvider" />
    ...
</bean>

Or by Spring version 3.1

<hz:hibernate-region-factory id="regionFactory" instance-ref="instance" />
...
<bean id="sessionFactory" class="org.springframework.orm.hibernate3.LocalSessionFactoryBean" scope="singleton">
    <property name="dataSource" ref="dataSource"/>
    <property name="cacheRegionFactory" ref="regionFactory" />
    ...
</bean>

14.5. Spring Data - JPA

Hazelcast supports JPA persistence integrated with Spring Data-JPA module. Your POJOs are mapped and persisted to your relational database. To use JPA persistence first you should create a Repository interface extending CrudRepository class with object type that you want to persist..

package com.hazelcast.jpa.repository;

import com.hazelcast.jpa.Product;
import org.springframework.data.repository.CrudRepository;

public interface ProductRepository extends CrudRepository<Product, Long> {

}

Then you should add your data source and repository definition to you Spring configuration,

<jpa:repositories
       base-package="com.hazelcast.jpa.repository" />

    <bean class="com.hazelcast.jpa.SpringJPAMapStore" id="jpamapstore">
        <property name="crudRepository" ref="productRepository" />
    </bean>

    <bean class="org.apache.commons.dbcp.BasicDataSource" destroy-method="close" id="dataSource">
        <property name="driverClassName" value="com.mysql.jdbc.Driver"/>
             <property name="url" value="jdbc:mysql://localhost:3306/YOUR_DB"/>
             <property name="username" value="YOUR_USERNAME"/>
             <property name="password" value="YOUR_PASSWORD"/>
    </bean>

    <bean id="entityManagerFactory"
      class="org.springframework.orm.jpa.LocalContainerEntityManagerFactoryBean">
      <property name="dataSource" ref="dataSource" />
      <property name="jpaVendorAdapter">
        <bean class="org.springframework.orm.jpa.vendor.HibernateJpaVendorAdapter">
          <property name="generateDdl" value="true" />
          <property name="database" value="MYSQL" />
        </bean>
      </property>
        <property name="persistenceUnitName" value="jpa.sample" />
    </bean>

    <bean class="org.springframework.orm.jpa.JpaTransactionManager"
      id="transactionManager">
      <property name="entityManagerFactory"
          ref="entityManagerFactory" />
      <property name="jpaDialect">
        <bean class="org.springframework.orm.jpa.vendor.HibernateJpaDialect" />
      </property>
    </bean>
    

In the example configuration above, Hibernate and MYSQL is configured, you change them according your ORM and database selection. Also you should define your persistence unit with persistence.xml under META-INF directory.

<?xml version="1.0" encoding="UTF-8"?>
<persistence version="2.0" xmlns="http://java.sun.com/xml/ns/persistence" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://java.sun.com/xml/ns/persistence http://java.sun.com/xml/ns/persistence/persistence_2_0.xsd">
	<persistence-unit name="jpa.sample" />
</persistence>

By default, the key is expected to be the same with id of the JPA object. You can change this behaviour and customize MapStore implementation extending SpringJPAMapStore class. For more info see Spring Data JPA Reference .

14.6. Spring Data - MongoDB

With version 2.1, Hazelcast will support MongoDB persistence integrated with Spring Data-MongoDB module. Spring MongoDB module maps your objects to equivalent MongoDB objects. To persist your objects into MongoDB you should define MongoDB mapstore in your Spring configuration as follows:

<mongo:mongo id="mongo" host="localhost" port="27017"/>

<bean id="mongoTemplate"
      class="org.springframework.data.mongodb.core.MongoTemplate">
    <constructor-arg ref="mongo"/>
    <constructor-arg name="databaseName" value="test"/>
</bean>

<bean class="com.hazelcast.spring.mongodb.MongoMapStore" id="mongomapstore">
    <property name="mongoTemplate" ref="mongoTemplate" />
</bean>

Then you can set this as mapstore for maps that you want to persist into MongoDB.

<hz:map name="user">
    <hz:map-store enabled="true" implementation="mongomapstore"
                  write-delay-seconds="0">
    </hz:map-store>
</hz:map>

By default, the key is set as id of the MongoDB object. You can override MongoMapStore class for you custom needs. For more info see Spring Data MongoDB Reference .

Chapter 15. Clients

There are currently three ways to connect to a running Hazelcast cluster:

15.1. Native Client

Native Client enables you to do all Hazelcast operations without being a member of the cluster. It connects to one of the cluster members and delegates all cluster wide operations to it. When the relied cluster member dies, client will transparently switch to another live member.

There can be hundreds, even thousands of clients connected to the cluster. But by default there is 40 threads on server side that will handle all the requests. You may want to increase hazelcast.executor.client.thread.count property for better performance.

Imagine a trading application where all the trading data stored and managed in a 10 node Hazelcast cluster. Swing/Web applications at traders' desktops can use Native Java Client to access and modify the data in the Hazelcast cluster.

Currently Hazelcast has Native Java and C# Client available.

LiteMember vs. Native Client

LiteMember is a member of the cluster, it has socket connection to every member in the cluster and it knows where the data is so it will get to the data much faster. But LiteMember has the clustering overhead and it must be on the same data center even on the same RAC. However Native client is not member and relies on one of the cluster members. Native Clients can be anywhere in the LAN or WAN. It scales much better and overhead is quite less. So if your clients are less than Hazelcast nodes then LiteMember can be an option; otherwise definitely try Native Client. As a rule of thumb: Try Native client first, if it doesn't perform well enough for you, then consider LiteMember.

The following picture describes the clients connecting to Hazelcast Cluster. Note the difference between LiteMember and Java Client

15.1.1. Java Client

You can do almost all hazelcast operations with Java Client. It already implements the same interface. You must include hazelcast-client.jar into your classpath.

import com.hazelcast.core.HazelcastInstance;
import com.hazelcast.client.HazelcastClient;

import java.util.Map;
import java.util.Collection;


ClientConfig clientConfig = new ClientConfig();
clientConfig.getGroupConfig().setName("dev").setPassword("dev-pass");
clientConfig.addAddress("10.90.0.1", "10.90.0.2:5702");

HazelcastInstance client = HazelcastClient.newHazelcastClient(clientConfig);
//All cluster operations that you can do with ordinary HazelcastInstance
Map<String, Customer> mapCustomers = client.getMap("customers");
mapCustomers.put("1", new Customer("Joe", "Smith"));
mapCustomers.put("2", new Customer("Ali", "Selam"));
mapCustomers.put("3", new Customer("Avi", "Noyan"));

Collection<Customer> colCustomers = mapCustomers.values();
for (Customer customer : colCustomers) {
     // process customer
}

15.1.2. CSharp Client (Enterprise Edition Only)

You can use native C# client to connect to the running Hazelcast instances. All you need is to include Hazelcast.Client.dll into your C# project. The API is very similar to Java native client. Note that C# client doesn't have automatic reconnection feature. If the node that it connected dies, it will not switch to another member. User must connect to another member itself.


using System;
using System.Collections.Generic;

using Hazelcast.Client;
using Hazelcast.Core;

ClientConfig clientConfig = new ClientConfig();
clientConfig.GroupConfig.Name = "dev";
clientConfig.GroupConfig.Password = "dev-pass";
clientConfig.addAddress("10.90.0.1");

HazelcastClient client = HazelcastClient.newHazelcastClient(clientConfig);
//Allmost all cluster operations that you can do with ordinary HazelcastInstance
//Note that the Customer class must have Serializable attribute or implement Hazelcast.IO.DataSerializable
IMap<String, Customer> mapCustomers = client.getMap("customers");
mapCustomers.put("1", new Customer("Joe", "Smith"));
mapCustomers.put("2", new Customer("Ali", "Selam"));
mapCustomers.put("3", new Customer("Avi", "Noyan"));

ICollection<Customer> colCustomers = mapCustomers.values();
foreach (Customer customer in colCustomers) {
     // process customer
}

You can serialize back and forth Java and C# Objects between C# client and Hazelcast server. All you need is to have your classes that you want to share to implement DataSerializable both on Java and C# in the exact same way. And on C# ClientConfig you must set a TypeConverter implementation that will convert Java Class name into C# Type and vice versa. A basic TypeConverter might look like this.


public class MyTypeConverter: Hazelcast.IO.ITypeConverter
{
    public string getJavaName(Type type)
    {
        if(type.Equals(typeof(Hazelcast.Client.Examples.MyCSharpClass)))
            return "com.hazelcast.examples.MyClass";

        return null;
    }

    public Type getType(String javaName)
    {
        if("com.hazelcast.examples.MyClass".Equals(javaName))
            return typeof(Hazelcast.Client.Examples.MyCSharpClass);

        return null;
    }
}

            

A basic MyCSharpClass implementing DataSerializable


using System;
using Hazelcast.IO;

public class MyCSharpClass: Hazelcast.IO.DataSerializable
{
    String field1 = "";
    int field2;

    public MyCSharpClass ()
    {
    }

    public MyCSharpClass (String f1, int f2)
    {
        this.field1 = f1;
        this.field2 = f2;
    }

    public void writeData(IDataOutput dout){
        dout.writeUTF(field1);
        dout.writeInt(field2);
    }

    public void readData(IDataInput din){
        field1 = din.readUTF();
        field2 = din.readInt();
    }
}

            

15.2. Memcache Client

A Memcache client written in any language can talk directly to Hazelcast cluster. No additional configuration is required. Here is an example: Let's say your cluster's members are:

Members [5] {
    Member [10.20.17.1:5701]
    Member [10.20.17.2:5701]
    Member [10.20.17.4:5701]
    Member [10.20.17.3:5701]
    Member [10.20.17.5:5701]
 }

And you have a PHP application that uses PHP Memcache client to cache things in Hazelcast. All you need to do is have your PHP memcache client connect to one of these members. It doesn't matter which member the client connects to because Hazelcast cluster looks as one giant machine (Single System Image). PHP client code sample:

<?php
    $memcache = new Memcache;
    $memcache->connect('10.20.17.1', 5701) or die ("Could not connect");
    $memcache->set('key1','value1',0,3600);
    $get_result = $memcache->get('key1'); //retrieve your data
    var_dump($get_result); //show it
?>

Notice that memcache client is connecting to 10.20.17.1 and using port5701. Java client code sample with SpyMemcached client:

MemcachedClient client = new MemcachedClient(AddrUtil.getAddresses("10.20.17.1:5701 10.20.17.2:5701"));
client.set("key1", 3600, "value1");
System.out.println(client.get("key1"));

An entry written with a memcache client can be read by another memcache client written in another language.

15.3. Rest Client

Let's say your cluster's members are:

Members [5] {
    Member [10.20.17.1:5701]
    Member [10.20.17.2:5701]
    Member [10.20.17.4:5701]
    Member [10.20.17.3:5701]
    Member [10.20.17.5:5701]
 }

And you have a distributed map named 'stocks'. You can put a new key1/value1 entry into this map by issuing HTTP POST call to http://10.20.17.1:5701/hazelcast/rest/maps/stocks/key1 URL. Your http post call's content body should contain the value (value1). You can retrieve this entry via HTTP GET call to http://10.20.17.1:5701/hazelcast/rest/maps/stocks/key1. You can also retrieve this entry from another member such ashttp://10.20.17.3:5701/hazelcast/rest/maps/stocks/key1.

RESTful access is provided through any member of your cluster. So you can even put an HTTP load-balancer in-front of your cluster members for load-balancing and fault-tolerance.

Now go ahead and install a REST plugin for your browser and explore further.

Hazelcast also stores the mime-type of your POST request if it contains any. So if, for example, you post binary of an image file and set the mime-type of the HTTP POST request to image/jpeg then this mime-type will be part of the response of your HTTP GET request for that entry.

Let's say you also have a task queue named 'tasks'. You can offer a new item into the queue via HTTP POST and take and item from the queue via HTTP DELETE.

HTTP POST http://10.20.17.1:5701/hazelcast/rest/queues/tasks <CONTENT> means

Hazelcast.getQueue("tasks").offer(<CONTENT>);

and HTTP DELETE http://10.20.17.1:5701/hazelcast/rest/queues/tasks/3 means

Hazelcast.getQueue("tasks").poll(3, SECONDS);

Note that you will have to handle the failures on REST polls as there is no transactional guarantee.

Chapter 16. Internals

16.1. Internals 2: Serialization

All your distributed objects such as your key and value objects, objects you offer into distributed queue and your distributed callable/runnable objects have to beSerializable.

Hazelcast serializes all your objects into an instance ofcom.hazelcast.nio.Data. Data is the binary representation of an object. When Hazelcast serializes an object intoData, it first checks whether the object is an instance of well-known, optimizable object such asString, Long, Integer, byte[], ByteBuffer, Date. If not, it then checks whether the object is an instance ofcom.hazelcast.nio.DataSerializable. If not, Hazelcast uses standard java serialization to convert the object into binary format. See com.hazelcast.nio.Serializer for details.

So for faster serialization, Hazelcast recommends to use of String, Long, Integer, byte[] objects or to implement com.hazelcast.nio.DataSerializable interface. Here is an example of a class implementing com.hazelcast.nio.DataSerializable interface.

public class Address implements com.hazelcast.nio.DataSerializable {
    private String street;
    private int zipCode;
    private String city;
    private String state;

    public Address() {}

    //getters setters..

    public void writeData(DataOutput out) throws IOException {
        out.writeUTF(street);
        out.writeInt(zipCode);
        out.writeUTF(city);
        out.writeUTF(state);
    }

    public void readData (DataInput in) throws IOException {
        street    = in.readUTF();
        zipCode = in.readInt();
        city    = in.readUTF();
        state    = in.readUTF();
    }
}

Lets take a look at another example which is encapsulating a DataSerializable field.

public class Employee implements com.hazelcast.nio.DataSerializable {
    private String firstName;
    private String lastName;
    private int age;
    private double salary;
    private Address address; //address itself is DataSerializable

    public Employee() {}

    //getters setters..

    public void writeData(DataOutput out) throws IOException {
        out.writeUTF(firstName);
        out.writeUTF(lastName);
        out.writeInt(age);
        out.writeDouble (salary);
        address.writeData (out);
    }

    public void readData (DataInput in) throws IOException {
        firstName = in.readUTF();
        lastName  = in.readUTF();
        age       = in.readInt();
        salary       = in.readDouble();
        address   = new Address();
        // since Address is DataSerializable let it read its own internal state
        address.readData (in);
    }
}

As you can see, since address field itself isDataSerializable, it is calling address.writeData(out) when writing and address.readData(in) when reading.

Caution: Hazelcast serialization is done on the user thread and it assumes that there will be only one object serialization at a time. So putting any Hazelcast operation that will require to serialize anything else will break the serialization. For Example: Putting

Hazelcast.getMap("anyMap").put("key", "dummy value");

line in readData or writeData methods will break the serialization. If you have to perform such an operation, at least it should be performed in another thread which will force the serialization to take on different thread.

16.2. Internals 3: Cluster Membership

It is important to note that Hazelcast is a peer to peer clustering so there is no 'master' kind of server in Hazelcast. Every member in the cluster is equal and has the same rights and responsibilities.

When a node starts up, it will check to see if there is already a cluster in the network. There are two ways to find this out:

  • Multicast discovery: If multicast discovery is enabled (that is the defualt) then the node will send a join request in the form of a multicast datagram packet.

  • Unicast discovery: if multicast discovery is disabled and TCP/IP join is enabled then the node will try to connect to he IPs defined in the hazelcast.xml configuration file. If it can successfully connect to at least one node, then it will send a join request through the TCP/IP connection.

If there is no existing node, then the node will be the first member of the cluster. If multicast is enabled then it will start a multicast listener so that it can respond to incoming join requests. Otherwise it will listen for join request coming viaTCP/IP.

If there is an existing cluster already, then the oldest member in the cluster will receive the join request and check if the request is for the right group. If so, the oldest member in the cluster will start the join process.

In the join process, the oldest member will:

  • send the new member list to all members

  • tell members to sync data in order to balance the data load

Every member in the cluster has the same member list in the same order. First member is the oldest member so if the oldest member dies, second member in the list becomes the first member in the list and the new oldest member.

See com.hazelcast.impl.Node and com.hazelcast.impl.ClusterManager for details.

Q. If, let say 50+, nodes are trying to join the cluster at the same time, are they going to join the cluster one by one?

No. As soon as the oldest member receives the first valid join request, it will wait 5 seconds for others to join so that it can join multiple members in one shot. If there is no new node willing to join for the next 5 seconds, then oldest member will start the join process. If a member leaves the cluster though, because of a JVM crash for example, cluster will immediately take action and oldest member will start the data recovery process.

16.3. Internals 4: Distributed Map

Hazelcast distributed map is a peer to peer, partitioned implementation so entries put into the map will be almost evenly partitioned onto the existing members. Entries are partitioned according to their keys.

Every key is owned by a member. So every key-aware operation, such as put, remove, get is routed to the member owning the key.

Q. How does Hazelcast determine the owner of a key?

Hazelcast creates fixed number of virtual partitions (blocks). Partition count is set to 271 by default. Each key falls into one of these partitions. Each partition is owned/managed by a member. Oldest member of the cluster will assign the ownerships of the partitions and let every member know who owns which partitions. So at any given time, each member knows the owner member of a each partition. Hazelcast will convert your key object to com.hazelcast.nio.Data then calculate the partition of the owner:partition-of-the-key = hash(keyData) % PARTITION_COUNT. Since each member(JVM) knows the owner of each partition, each member can find out which member owns the key.

Q. Can I get the owner of a key?

Yes. Use Partition API to get the partition that your key falls into and then get the owner of that partition. Note that owner of the partition can change over time as new members join or existing members leave the cluster.

PartitionService partitionService = Hazelcast.getPartitionService();
Partition partition = partitionService.getPartition(key);
Member ownerMember = partition.getOwner();

Locally owned entries can be obtained by callingmap.localKeySet().

Q. What happens when a new member joins?

Just like any other member in the cluster, the oldest member also knows who owns which partition and what the oldest member knows is always right. The oldest member is also responsible for redistributing the partition ownerships when a new member joins. Since there is new member, oldest member will take ownership of some of the partitions and give them to the new member. It will try to move the least amount of data possible. New ownership information of all partitions is then sent to all members.

Notice that the new ownership information may not reach each member at the same time and the cluster never stops responding to user map operations even during joins so if a member routes the operation to a wrong member, target member will tell the caller to re-do the operation.

If a member's partition is given to the new member, then the member will send all entries of that partition to the new member (Migrating the entries). Eventually every member in the cluster will own almost same number of partitions, and almost same number of entries. Also eventually every member will know the owner of each partition (and each key).

You can listen for migration events. MigrationEvent contains thepartitionId,oldOwner, and newOwner information.

PartitionService partitionService = Hazelcast.getPartitionService();
partitionService.addMigrationListener(new MigrationListener () {

   public void migrationStarted(MigrationEvent migrationEvent) {
      System.out.println(migrationEvent);
   }

   public void migrationCompleted(MigrationEvent migrationEvent) {
      System.out.println(migrationEvent);
   }
});

Q. How about distributed set and list?

Both distributed set and list are implemented on top of distributed map. The underlying distributed map doesn't hold value; it only knows the key. Items added to both list and set are treated as keys. Unlike distributed set, since distributed list can have duplicate items, if an existing item is added again, copyCount of the entry (com.hazelcast.impl.ConcurrentMapManager.Record) is incremented. Also note that index based methods of distributed list, such as List.get(index) andList.indexOf(Object), are not supported because it is too costly to keep distributed indexes of list items so it is not worth implementing.

Check out the com.hazelcast.impl.ConcurrentMapManager class for the implementation. As you will see, the implementation is lock-free because ConcurrentMapManager is a singleton and processed by only one thread, theServiceThread.

Chapter 17. Management Center

17.1. Introduction

Hazelcast Management Center enables you to monitor and manage your servers running hazelcast. With Management Center, in addition to monitoring overall state of your clusters, you can also analyze and browse your data structures in details. You can also update map configurations and take thread dump from nodes. With its scripting module, you can run scritps (JavaScript, Groovy etc.) on your servers. Version 2.0 is a web based tool so you can deploy it into your internal server and serve your users.

17.1.1. Installation

It is important to understand how it actually works. Basically you will deploy mancenter.war application into your Java web server and then tell Hazelcast nodes to talk to that web application. That means, your Hazelcast nodes should know the URL of mancenter application before they start.

Here are the steps:

  • Download the latest Hazelcast zip from hazelcast.com

  • Zip contains mancenter.war file. Deploy it to your web server (Tomcat, Jetty etc.) Let's say it is running athttp://localhost:8080/mancenter.

  • Start your web server and make sure http://localhost:8080/mancenter is up.

  • Configure your Hazelcast nodes by adding the URL of your web app to your hazelcast.xml. Hazelcast nodes will send their states to this URL.

    <management-center enabled="true">http://localhost:8080/mancenter</management-center>

  • Start your hazelcast cluster.

  • Browse to http://localhost:8080/mancenter and login. Initial login username/passwords is admin/admin

Management Center creates a directory with name "mancenter" under your "user/home" directory to save data files. You can change the data directory setting "hazelcast.mancenter.home" system property.

17.1.2. User Administration

Default credentials are for the admin user. In the Administration tab, Admin can add/remove/update users and control user read/write permissions.

17.1.3. Tool Overview

The starter page of the tool isCluster Home. Here you can see cluster's main properties such as uptime, memory. Also with pie chart, you can see the distribution of partitions over cluster members. You can come back to this page, by clicking the Home icon on the top-right toolbar. On the left panel you see the Map/Queue/Topic instances in the cluster. At the bottom-left corner, members of the cluster are listed. On top menu bar, you can change the current tab toScripting, Docs, userAdministration. Note that Administration tab is viewable only for admin users. Also Scripting page is disabled for users with read-only credential.

17.2. Maps

Map instances are listed on the left panel. When you click on a map, a new tab for monitoring this map instance is opened on the right. In this tab, you can monitor metrics also re-configure the map.

17.2.1. Monitoring Maps

In map page you can monitor instance metrics by 2 charts and 2 datatables. First data table "Memory Data Table" gives the memory metrics distributed over members. "Throughput Data Table" gives information about the operations performed on instance (get, put, remove) Each chart monitors a type data of the instance on cluster. You can change the type by clicking on chart. The possible ones are: Size, Throughput, Memory, Backup Size, Backup Memory, Hits, Locked Entries, Puts, Gets, Removes...

17.2.2. Map Browser

You can open "Map Browser" tool by clicking "Browse" button on map tab page. Using map browser, you can reach map's entries by keys. Besides its value, extra informations such as entry's cost, expiration time is provided.

17.2.3. Map Configuration

You can open "Map Configuration" tool by clicking "Config" button on map tab page. This button is disabled for users with Read-Only permission. Using map config tool you can adjust map's setting. You can change backup count, max size, max idle(seconds), eviction policy, cache value, read backup data, backup count of the map.

17.3. Queues

Queues is the second data structure that you can monitor in management center. You can activate the Queue Tab by clicking the instance name listed on the left panel under queues part. The queue page consists of the charts monitoring data about the queue. You can change the data to be monitored by clicking on the chart. Available options are Size, Polls, Offers.

17.4. Topics

You can monitor your topics' metrics by clicking the topic name listed on the left panel under topics part. There are two charts which reflects live data, and a datatable lists the live data distributed among members.

17.5. Members

The current members in the cluster are listed on the bottom side of the left panel. You can monitor each member on tab page displayed by clicking on member items.

17.5.1. Monitoring

In members page there are 4 inner tab pages to monitor meber's state and properties. Runtime: Runtime properties about memory, threads are given. This data updates dynamically. Properties: System properties are displayed. Configuration: Configuration xml initially set can be viewed here. Partitions: The partitions belongs to this member are listed.

17.5.2. Operations

Besides monitoring you can perform certain actions on members. You can take thread dump of the member and you can perform garbage collection on the selected member.

17.6. System Logs

System logs part helps you track internal operations and detect problems. To see system logs first you should set a Log level other than "None". In left menu there are inputs by which you can filter the displayed logs dynamically. Also you can Export your logs and send the file to Hazelcast support team, so they can analyze and help you solving your problem.

17.7. Scripting

In scripting part, you can execute your own code on your cluster. In the left part you can select members, on which the code will be executed. Also you can select over scripting languages: Javascript, Groovy, JRuby, BeanShell. This part is only enabled for users with read/write permissions for current cluster.

17.8. Time Travel

Time Travel mode is activated by clicking clock icon on top right toolbar. In time travel mode, the time is paused and the full state of the cluster is displayed according the time selected on time slider. You can change time either by Prev/Next buttons or sliding the slider. Also you can change the day by clicking calendar icon. Management center stores the states in you local disk, while your web server is alive. So if you slide to a time when you do not have data, the reports will be seen as empty.

17.9. Console

The console tool enables you execute commands on the cluster. You can read or write on instances but first you should set namespace. For example if you have a map with name "mapCustomers". To get a customer with key "Jack" you should first set the namespace with command "ns mapCustomers". Then you can take the object by "m.get Jack" Here is the command list:

-- General commands
echo true|false                      //turns on/off echo of commands (default false)
silent true|false                    //turns on/off silent of command output (default false)
#<number> <command>                  //repeats <number> time <command>, replace $i in <command> with current iteration (0..<number-1>)
&<number> <command>                  //forks <number> threads to execute <command>, replace $t in <command> with current thread number (0..<number-1>
     When using #x or &x, is is advised to use silent true as well.
     When using &x with m.putmany and m.removemany, each thread will get a different share of keys unless a start key index is specified
jvm                                  //displays info about the runtime
who                                  //displays info about the cluster
whoami                               //displays info about this cluster member
ns <string>                          //switch the namespace for using the distributed queue/map/set/list <string> (defaults to "default"
@<file>                              //executes the given <file> script. Use '//' for comments in the script

-- Queue commands
q.offer <string>                     //adds a string object to the queue
q.poll                               //takes an object from the queue
q.offermany <number> [<size>]        //adds indicated number of string objects to the queue ('obj<i>' or byte[<size>])
q.pollmany <number>                  //takes indicated number of objects from the queue
q.iterator [remove]                  //iterates the queue, remove if specified
q.size                               //size of the queue
q.clear                              //clears the queue

-- Set commands
s.add <string>                       //adds a string object to the set
s.remove <string>                    //removes the string object from the set
s.addmany <number>                   //adds indicated number of string objects to the set ('obj<i>')
s.removemany <number>                //takes indicated number of objects from the set
s.iterator [remove]                  //iterates the set, removes if specified
s.size                               //size of the set
s.clear                              //clears the set

-- Lock commands
lock <key>                           //same as Hazelcast.getLock(key).lock()
tryLock <key>                        //same as Hazelcast.getLock(key).tryLock()
tryLock <key> <time>                 //same as tryLock <key> with timeout in seconds
unlock <key>                         //same as Hazelcast.getLock(key).unlock()

-- Map commands
m.put <key> <value>                  //puts an entry to the map
m.remove <key>                       //removes the entry of given key from the map
m.get <key>                          //returns the value of given key from the map
m.putmany <number> [<size>] [<index>]//puts indicated number of entries to the map ('key<i>':byte[<size>], <index>+(0..<number>)
m.removemany <number> [<index>]      //removes indicated number of entries from the map ('key<i>', <index>+(0..<number>)
     When using &x with m.putmany and m.removemany, each thread will get a different share of keys unless a start key <index> is specified
m.keys                               //iterates the keys of the map
m.values                             //iterates the values of the map
m.entries                            //iterates the entries of the map
m.iterator [remove]                  //iterates the keys of the map, remove if specified
m.size                               //size of the map
m.clear                              //clears the map
m.destroy                            //destroys the map
m.lock <key>                         //locks the key
m.tryLock <key>                      //tries to lock the key and returns immediately
m.tryLock <key> <time>               //tries to lock the key within given seconds
m.unlock <key>                       //unlocks the key

-- List commands:
l.add <string>
l.add <index> <string>
l.contains <string>
l.remove <string>
l.remove <index>
l.set <index> <string>
l.iterator [remove]
l.size
l.clear
-- AtomicNumber commands:
a.get
a.set <long>
a.inc
a.dec
-- Executor Service commands:
execute	<echo-input>				//executes an echo task on random member
execute0nKey	<echo-input> <key>		//executes an echo task on the member that owns the given key
execute0nMember <echo-input> <key>	//executes an echo task on the member with given index
execute0nMembers <echo-input> 		//executes an echo task on all of the members

Chapter 18. Miscellaneous

18.1. Common Gotchas

Hazelcast is the distributed implementation of several structures that exist in Java. Most of the time it behaves as you expect. However there are some design choices in Hazelcast that violate some contracts. This page will list those violations.

  1. equals() and hashCode() methods for the objects stored in Hazelcast

    When you store a key, value in a distributed Map, Hazelcast serializes the key and value and stores the byte array version of them in local ConcurrentHashMaps. And this ConcurrentHashMap uses the equals and hashCode methods of byte array version of your key. So it does not take into account the actual equals and hashCode implementations of your objects. So it is important that you choose your keys in a proper way. Implementing the equals and hashCode is not enough, it is also important that the object is always serialized into the same byte array. All primitive types, like; String, Long, Integer and etc. are good candidates for keys to use in Hazelcast. An unsorted Set is an example of a very bad candidate because Java Serialization may serialize the same unsorted set in two different byte arrays.

    Note that the distributed Set and List stores its entries as the keys in a distributed Map. So the notes above apply to the objects you store in Set and List.

  2. Hazelcast always return a clone copy of a value. Modifying the returned value does not change the actual value in the map (or multimap or list or set). You should put modified value back to make changes visible to all nodes.

    V value = map.get(key);
    value.updateSomeProperty();
    map.put(key, value);
    

    If cache-value is true (default is true), Hazelcast caches that returned value for fast access in local node. Modifications done to this cached value without putting it back to map will be visible to only local node, successive get calls will return the same cached value. To reflect modifications to distributed map, you should put modified value back into map.

  3. Collections which return values of methods such as IMap.keySet, IMap.values, IMap.entrySet, MultiMap.get, MultiMap.remove, IMap.keySet, IMap.values, contain cloned values. These collections are NOT backup by related Hazelcast objects. So changes to the these are NOT reflected in the originals, and vice-versa.

  4. Most of the Hazelcast operations throw an RuntimeInterruptedException (which is unchecked version of InterruptedException) if a user thread is interrupted while waiting a response. Hazelcast uses RuntimeInterruptedException to pass InterruptedException up through interfaces that don't have InterruptedException in their signatures. Users should be able to catch and handle RuntimeInterruptedException in such cases as if their threads are interrupted on a blocking operation.

  5. Some of Hazelcast operations can throw ConcurrentModificationException under transaction while trying to acquire a resource, although operation signatures don't define such an exception. Exception is thrown if resource can not be acquired in a specific time. Users should be able to catch and handle ConcurrentModificationException while they are using Hazelcast transactions.

18.2. Testing Cluster

Hazelcast allows you to create more than one member on the same JVM. Each member is called HazelcastInstance and each will have its own configuration, socket and threads, so you can treat them as totally separate members. This enables us to write and run cluster unit tests on single JVM. As you can use this feature for creating separate members different applications running on the same JVM (imagine running multiple webapps on the same JVM), you can also use this feature for testing Hazelcast cluster.

Let's say you want to test if two members have the same size of a map.

@Test
public void testTwoMemberMapSizes() {
    // start the first member
    HazelcastInstance h1 = Hazelcast.newHazelcastInstance(null);
    // get the map and put 1000 entries
    Map map1 = h1.getMap("testmap");
    for (int i = 0; i < 1000; i++) {
        map1.put(i, "value" + i);
    }
    // check the map size
    assertEquals(1000, map1.size());
    // start the second member
    HazelcastInstance h2 = Hazelcast.newHazelcastInstance(null);
    // get the same map from the second member
    Map map2 = h2.getMap("testmap");
    // check the size of map2
    assertEquals(1000, map2.size());
    // check the size of map1 again
    assertEquals(1000, map1.size());
}

In the test above, everything happened in the same thread. When developing multi-threaded test, coordination of the thread executions has to be carefully handled. Usage of CountDownLatch for thread coordination is highly recommended. You can certainly use other things. Here is an example where we need to listen for messages and make sure that we got these messages:

@Test
public void testTopic() {
    // start two member cluster
    HazelcastInstance h1 = Hazelcast.newHazelcastInstance(null);
    HazelcastInstance h2 = Hazelcast.newHazelcastInstance(null);
    String topicName = "TestMessages";
    // get a topic from the first member and add a messageListener
    ITopic<String> topic1 = h1.getTopic(topicName);
    final CountDownLatch latch1 = new CountDownLatch(1);
    topic1.addMessageListener(new MessageListener() {
        public void onMessage(Object msg) {
            assertEquals("Test1", msg);
            latch1.countDown();
        }
    });
    // get a topic from the second member and add a messageListener
    ITopic<String> topic2 = h2.getTopic(topicName);
    final CountDownLatch latch2 = new CountDownLatch(2);
    topic2.addMessageListener(new MessageListener() {
        public void onMessage(Object msg) {
            assertEquals("Test1", msg);
            latch2.countDown();
        }
    });
    // publish the first message, both should receive this
    topic1.publish("Test1");
    // shutdown the first member
    h1.shutdown();
    // publish the second message, second member's topic should receive this
    topic2.publish("Test1");
    try {
        // assert that the first member's topic got the message
        assertTrue(latch1.await(5, TimeUnit.SECONDS));
        // assert that the second members' topic got two messages
        assertTrue(latch2.await(5, TimeUnit.SECONDS));
    } catch (InterruptedException ignored) {
    }
}

You can surely start Hazelcast members with different configuration. Let's say we want to test if Hazelcast LiteMember can shutdown fine.

@Test(timeout = 60000)
public void shutdownLiteMember() {
    // first config for normal cluster member
    Config c1 = new XmlConfigBuilder().build();
    c1.setPortAutoIncrement(false);
    c1.setPort(5709);
    // second config for LiteMember
    Config c2 = new XmlConfigBuilder().build();
    c2.setPortAutoIncrement(false);
    c2.setPort(5710);
    // make sure to set LiteMember=true
    c2.setLiteMember(true);
    // start the normal member with c1
    HazelcastInstance hNormal = Hazelcast.newHazelcastInstance(c1);
    // start the LiteMember with different configuration c2
    HazelcastInstance hLite = Hazelcast.newHazelcastInstance(c2);
    hNormal.getMap("default").put("1", "first");
    assert hLite.getMap("default").get("1").equals("first");
    hNormal.shutdown();
    hLite.shutdown();
}

Also remember to call Hazelcast.shutdownAll() after each test case to make sure that there is no other running member left from the previous tests.

@After
public void cleanup() throws Exception {
    Hazelcast.shutdownAll();
}

Need more info? Check out existing tests.

18.3. Planned Features

Random order of planned features.

  • Native C++ Client

  • Ready-to-go Hazelcast Cache Server Image for Amazon EC2

  • Symmetric Encryption and SSL support for Java Client

  • Continuous query (events based on given criteria)

  • Distributed java.util.concurrent.DelayQueue implementation.

  • Cluster-wide receive ordering for topics.

  • Distributed Tree implementation.

  • Distributed Tuple implementation.

  • Call interceptors for modifying the request or the response.

  • Built-in file based storage.

History of existing features is available atRelease Notes.

18.4. Release Notes

Please see, Todo page for planned features.

2.3.1

2.3

  • Changed hazelcast.max.operation.timeout unit from seconds to milliseconds.

  • Added hazelcast.max.concurrent.operation.limit parameter to be able to limit number of concurrent operations can be submitted to Hazelcast.

  • Added hazelcast.backup.redo.enabled parameter to enable/disable redo for backup operations.

  • Added MultiMap and Distributed ExecutorService statistics to Management Center application.

  • MigrationListener has now an additional method to receive failed migration events; void migrationFailed(MigrationEvent migrationEvent).

  • ItemEvent has now an additional method returning Member firing that event; public Member getMember().

  • Improved out of memory (OOM) error detection and handling. Now it is possible to register custom hook when OutOfMemoryError is thrown. See Hazelcast.setOutOfMemoryHandler(OutOfMemoryHandler) and OutOfMemoryHandler javadocs for additional details.

  • Fixed some issues related to domain name handling and networking/join.

  • During cluster merge after a network split-brain, merging side will now fire MERGING and MERGED before and after RESTARTING and RESTARTED LifecycleEvents.

  • Fixed issues:

    167, 182, 188, 209, 217, 218, 220, 222, 223, 225, 226, 227, 228, 229, 230, 233, 234, 235, 237, 241, 243, 244, 247, 250

2.2

2.1.3

  • Fixed join and split brain issues. Improved domain name handling.

  • Fixed host aware partition group when all members are lite but one.

  • Fixed jmx management service dead-lock issue.

  • Fixed IMap replace-if-same and and remove-if-same concurrency issues.

  • Fixed issues:

    176, 177, 179, 181, 183, 185, 186, 189, 195, 196, 198

2.1.2

  • Fixed interruption handling to avoid dead-locks and inconsistency. See configuration property hazelcast.force.throw.interrupted.exception in ConfigurationProperties page.

  • Replaced ExecutorService used by migration process with a custom one.

  • Fixed ExecutorService a memory leak during executor thread termination.

  • Fixed issues:

    161, 174

2.1.1

2.1

  • IPv6 Support: Hazelcast now supports IPv6 addresses seamlessly.

  • Async backup support.

  • Spring Cache support: Hazelcast can be used as Spring Cache provider.

  • Spring Dependency Injection support: Hazelcast can apply bean properties or to apply factory callbacks such as ApplicationContextAware, BeanNameAware or to apply bean post-processing such as InitializingBean, @PostConstruct like annotations while using Hazelcast distributed ExecutorService or DistributedTasks or more generally any Hazelcast managed object.

  • Persistence support with Spring-Data MongoDB and JPA integration.

  • Member.getUuid() will return UUID for node.

  • Improved session clustering configuration.

  • Fixed issues:

    • Google Code

      809, 811, 816, 818

    • Github

      6, 7, 92, 98, 99, 101, 102, 103, 104, 108, 109, 110, 114, 116, 117, 118, 119, 120, 126, 127, 128, 130, 131, 132, 134, 135

2.0

  • New Elastic Memory(Enterprise Edition Only): By default, Hazelcast stores your distributed data (map entries, queue items) into Java heap which is subject to garbage collection. As your heap gets bigger, garbage collection might cause your application to pause tens of seconds, badly effecting your application performance and response times. Elastic Memory is Hazelcast with off-heap memory storage to avoid GC pauses. Even if you have terabytes of cache in-memory with lots of updates, GC will have almost no effect; resulting in more predictable latency and throughput.

  • Security Framework(Enterprise Edition Only): Hazelcast Security is JAAS based pluggable security framework which can be used to authenticate both cluster members and clients and do access control checks on client operations. With the security framework, take control of who can be part of the cluster or connect as client and which operations are allowed or not.

  • Native C# Client(Enterprise Edition Only): Just like our Native Java Client, it supports all map, multimap, queue, topic operations including listeners and queries.

  • Distributed Backups: Data owned by a member will be evenly backed up by all the other members. In other word, every member takes equal responsibility to backup every other node. This leads to better memory usage and less influence in the cluster when you add/remove nodes. The new backup system makes it possible to form backup-groups so that you can have backups and owners fall into different groups.

  • Parallel IO: Number of socket selector threads can be configured. You can have more IO threads, if you have good number of CPU/cores and high-throughput network.

  • Connection Management: Hazelcast 2.0 is more tolerant to connection failures. On connection failure it tries to repair it before declaring the member as dead. So now it is ok to have short socket disconnections… No problem if your virtual server migrates to a new host.

  • Listeners such as migration, membership and map indexes can be added with configuration.

  • New Event Objects: Event Listeners for Queue/List/Set/Topic were delivering the item itself on event methods. That’s why the items had to be deserialized by Hazelcast Threads before invoking the listeners. Sometimes this was causing class loader problems too. With 2.0, we have introduced new event containers for Queue/List/Set and Topic just like Map has EntryEvent. The new listeners now receive ItemEvent and Message objects respectively. The actual items are deserialized only if you call the appropriate get method on the event objects. This is where we brake the compatibility with the older versions of Hazelcast.

  • ClientConfig API: We had too many of factory methods to instantiate a HazelcastClient. Now all we need isHazelcastClient.newHazelcastClient(ClientConfig).

  • SSL communication support among cluster nodes.

  • Distributed MultiMap value collection can be either List or Set.

  • SuperClient is renamed to LiteMember to avoid confusion. Be careful! It is a member, not a client.

  • New IMap.set (key, value, ttl, TimeUnit) implementation, which is optimized put(key, value) operation as set doesn’t return the old value.

  • HazelcastInstance.getLifecycleService().kill() will forcefully kill the node. Useful for testing.

  • forceUnlock, to unlock the locked entry from any node and any thread regardless of the owner.

  • Enum type query support.. new SqlPredicate (“level = Level.WARNING”) for example

  • Fixed issues: 430, 459, 471, 567, 574, 582, 629, 632, 646, 666, 686, 669, 690, 692, 693, 695, 698, 705, 708, 710, 711, 712, 713, 714, 715, 719 , 721, 722, 724, 727, 728, 729, 730, 731, 732, 733, 735, 738, 739, 740, 741, 742, 747, 751, 752, 754, 756, 758, 759, 760, 761, 765, 767, 770, 773, 779, 781, 782, 783, 787, 790, 795, 796

1.9.4

  • New WAN Replication (synchronization of separate active clusters)

  • New Data Affinity (co-location of related entries) feature.

  • New EC2 Auto Discovery for your Hazelcast cluster running on Amazon EC2 platform.

  • New Distributed CountDownLatch implementation.

  • New Distributed Semaphore implementation.

  • Improvement: Distribution contains HTML and PDF documentation besides Javadoc.

  • Improvement: Better TCP/IP and multicast join support. Handling more edge cases like multiple nodes starting at the same time.

  • Improvement: Memcache protocol: Better integration between Java and Memcache clients. Put from memcache, get from Java client.

  • Monitoring Tool is removed from the project.

  • 200+ commits 25+ bug fixes and several other enhancements.

1.9.3

  • Re-implementation of distributed queue.

    • Configurable backup-count and synchronous backup.

    • Persistence support based on backing MapStore

    • Auto-recovery from backing MapStore on startup.

  • Re-implementation of distributed list supporting index based operations.

  • New distributed semaphore implementation.

  • Optimized IMap.putAll for much faster bulk writes.

  • New IMap.getAll for bulk reads which is calling MapLoader.loadAll if necessary.

  • New IMap.tryLockAndGet and IMap.putAndUnlock API

  • New IMap.putTransient API for storing only in-memory.

  • New IMap.addLocalEntryListener() for listening locally owned entry events.

  • New IMap.flush() for flushing the dirty entries into MapStore.

  • New MapLoader.getAllKeys API for auto-pre-populating the map when cluster starts.

  • Support for min. initial cluster size to enable equally partitioned start.

  • Graceful shutdown.

  • Faster dead-member detection.

1.9

  • Memcache interface support. Memcache clients written in any language can access Hazelcast cluster.

  • RESTful access support. http://<ip>:5701/hazelcast/rest/maps/mymap/key1

  • Split-brain (network partitioning) handling

  • New LifecycleService API to restart, pause Hazelcast instances and listen for the lifecycle events.

  • New asynchronous put and get support for IMap via IMap.asyncPut() and IMap.asyncGet()

  • New AtomicNumber API; distributed implementation of java.util.concurrent.atomic.AtomicLong

  • So many bug fixes.

1.8.4

  • Significant performance gain for multi-core servers. Higher CPU utilization and lower latency.

  • Reduced the cost of map entries by 50%.

  • Better thread management. No more idle threads.

  • Queue Statistics API and the queue statistics panel on the Monitoring Tool.

  • Monitoring Tool enhancements. More responsive and robust.

  • Distribution contains hazelcast-all-<version>.jar to simplify jar dependency.

  • So many bug fixes.

1.8.3

  • Bug fixes

  • Sorted index optimization for map queries.

1.8.2

  • A major bug fix

  • Minor optimizations

1.8.1

  • Hazelcast Cluster Monitoring Tool (see the hazelcast-monitor-1.8.1.war in the distro)

  • New Partition API. Partition and key owner, migration listeners.

  • New IMap.lockMap() API.

  • New Multicast+TCP/IP join feature. Try multicast first, if not found, try tcp/ip.

  • New Hazelcast.getExecutorService(name) API. Have separate named ExecutorServices. Do not let your big tasks blocking your small ones.

  • New Logging API. Build your own logging. or simply use Log4j or get logs as LogEvents.

  • New MapStatistics API. Get statistics for your Map operations and entries.

  • HazelcastClient automatically updates the member list. no need to pass all members.

  • Ability to start the cluster members evenly partitioned. so no migration.

  • So many bug fixes and enhancements.

  • There are some minor Config API change. Just make sure to re-compile.

1.8

  • Java clients for accessing the cluster remotely. (C# is next)

  • Distributed Query for maps. Both Criteria API and SQL support.

  • Near cache for distributed maps.

  • TTL (time-to-live) for each individual map entry.

  • IMap.put(key,value, ttl, timeunit)

  • IMap.putIfAbsent(key,value, ttl, timeunit)

  • Many bug fixes.

1.7.1

  • Multiple Hazelcast members on the same JVM. New HazelcastInstance API.

  • Better API based configuration support.

  • Many performance optimizations. Fastest Hazelcast ever!

  • Smoother data migration enables better response times during joins.

  • Many bug fixes.

1.7

  • Persistence via Loader/Store interface for distributed map.

  • Socket level encryption. Both symmetric and asymmetric encryption supported.

  • New JMX support. (many thanks to Marco)

  • New Hibernate second level cache provider (many thanks to Leo)

  • Instance events for getting notified when a data structure instance (map, queue, topic etc.) is created or destroyed.

  • Eviction listener. EntryListener.entryEvicted(EntryEvent)

  • Fully 'maven'ized.

  • Modularized...

  • hazelcast (core library)

  • hazelcast-wm (http session clustering tool)

  • hazelcast-ra (JCA adaptor)

  • hazelcast-hibernate (hibernate cache provider)

1.6

  • Support for synchronous backups and configurable backup-count for maps.

  • Eviction support. Timed eviction for queues. LRU, LFU and time based eviction for maps.

  • Statistics/history for entries. create/update time, number of hits, cost. see IMap.getMapEntry(key)

  • MultiMap implementation. similar to google-collections and apache-common-collections MultiMap but distributed and thread-safe.

  • Being able to destroy() the data structures when not needed anymore.

  • Being able to Hazelcast.shutdown() the local member.

  • Get the list of all data structure instances viaHazelcast.getInstances().

1.5

  • Major internal refactoring

  • Full implementation ofjava.util.concurrent.BlockingQueue. Now queues can have configurable capacity limits.

  • Super Clients (a.k.a LiteMember): Members with no storage. If -Dhazelcast.super.client=true JVM parameter is set, that JVM will join the cluster as a 'super client' which will not be a 'data partition' (no data on that node) but will have super fast access to the cluster just like any regular member does.

  • Http Session sharing support for Hazelcast Web Manager. Different webapps can share the same sessions.

  • Ability to separate clusters by creating groups. ConfigGroup

  • java.util.logging support.

1.4

  • Add, remove and update events for queue, map, set and list

  • Distributed Topic for pub/sub messaging

  • Integration with J2EE transactions via JCA complaint resource adapter

  • ExecutionCallback interface for distributed tasks

  • Cluster-wide unique id generator

1.3

  • Transactional Distributed Queue, Map, Set and List

1.2

  • Distributed Executor Service

  • Multi member executions

  • Key based execution routing

  • Task cancellation support

1.1

  • Session Clustering with Hazelcast Webapp Manager

  • Full TCP/IP clustering support

1.0

  • Distributed implementation of java.util.{Queue,Map,Set,List}

  • Distributed implementation of java.util.concurrency.Lock

  • Cluster Membership Events