Book Image

Getting Started with Hazelcast, Second Edition

By : Matthew Johns
Book Image

Getting Started with Hazelcast, Second Edition

By: Matthew Johns

Overview of this book

Related Content you might be interested in

Current Title:

Small book image
Getting Started with Hazelcast, Second Edition
Matthew Johns
Jul, 2015 | 162 pages
No titles found
Table of Contents (19 chapters)
Getting Started with Hazelcast Second Edition
Getting Started with Hazelcast Second Edition
Credits
Credits
About the Author
About the Author
About the Reviewers
About the Reviewers
www.PacktPub.com
www.PacktPub.com
Preface
Preface
Free Chapter
1
What is Hazelcast?
What is Hazelcast?
Starting out as usual
Data deciding to hang around
Therein lies the problem
Breaking the mould
Moving to new ground
Summary
2
Getting off the Ground
Getting off the Ground
Let's get started
Showing off straightaway
Mapping back to the real world
Sets, lists, and queues
Many things at a time
Searching and indexing
What happens when we reach our limits?
Summary
3
Going Concurrent
Going Concurrent
Atomic control
Distributed locking
Transactionally rolling on
Spreading the word
Summary
4
Divide and Conquer
Divide and Conquer
Divvying up the data
Backups everywhere and nowhere
Scaling up the cluster
Having some of our data everywhere
Grouping and separating nodes
Network partitioning
Maintaining quorum
Summary
5
Listening Out
Listening Out
Listening to the goings-on
The sound of our own data
Programmatic configuration ahead of time
Events unfolding in the wider world
Moving data around the place
Extending quorum
Summary
6
Spreading the Load
Spreading the Load
All power to the compute
Running once, running everywhere
Placing tasks next to the data
Summary
7
Gathering Results
Gathering Results
What is this big data hype all about?
Putting theory into practice
Simplifying just aggregating up
Summary
8
Typical Deployments
Typical Deployments
All heap and nowhere to go
Stepping back from the cluster
Serialization and classes
Getting straight to the point
Architectural overview
Summary
9
From the Outside Looking In
From the Outside Looking In
What about the rest of us?
Memcache
Going RESTful
Summary
10
Going Global
Going Global
Getting set up in the cloud
Spreading out around the globe
Summary
11
Playing Well with Others
Playing Well with Others
Don't pass what you need, depend on it
Transparently caching others' data
Cacheable methods with the Spring cache
Caching by standard
Collection persistence
Web session storage
Cluster management
Summary
Configuration Summary
Configuration Summary
Cluster name
Management center
General properties
WAN replication configuration
Local network/interface binding
Cluster discovery (multicast-, unicast-, or EC2-based)
Map configuration
Listeners notified on map events
Listeners notified on topic broadcast
Cluster-wide listener registration
Index
Index

Map configuration

This is used to configure how map data is stored within the cluster, either individual named maps or a cluster-wide default.

Maximum size and eviction

Maximum size policies include:

  • PER_NODE

  • PER_PARTITION

  • USED_HEAP_SIZE

  • USED_HEAP_PERCENTAGE

  • FREE_HEAP_SIZE

  • FREE_HEAP_PERCENTAGE

Eviction policies include:

  • LRU: Least Recently Used

  • LFU: Least Frequently Used

  • NONE

<map name="default">
 <max-size policy="PER_NODE">10</max-size>
 <eviction-policy>LFU</eviction-policy>
 <eviction-percentage>20</eviction-percentage>
</map>

config.getMapConfig("default")
 .setMaxSizeConfig(new MaxSizeConfig(10, MaxSizePolicy.PER_NODE))
 .setEvictionPolicy(EvictionPolicy.LFU)
 .setEvictionPercentage(20);

Backup copies

Backup = Synchronous copies

Async = Asynchronous copies

Total copies = 1 Main + Backup + Async

<map name="default">
 <backup-count>1</backup-count>
 <async-backup-count>1</async-backup-count>
</map>

config.getMapConfig("default")
 .setBackupCount(1)
 .setAsyncBackupCount(1);

Age expiry and idle expiry times

This configures when entries will be by default automatically removed:

<map name="default">
 <time-to-live-seconds>86400</time-to-live-seconds>
 <max-idle-seconds>3600</max-idle-seconds>
</map>

config.getMapConfig("default")
 .setTimeToLiveSeconds(86400)
 .setMaxIdleSeconds(3600);

Map merge policy

Default merge policy classes

  • com.hazelcast.map.merge.PassThroughMergePolicy

  • com.hazelcast.map.merge.PutIfAbsentMapMergePolicy

  • com.hazelcast.map.merge.HigherHitsMapMergePolicy

  • com.hazelcast.map.merge.LatestUpdateMapMergePolicy 

    <map name="default">
 <merge-policy>
  com.hazelcast.map.merge.LatestUpdateMapMergePolicy
 </merge-policy>
</map>

config.getMapConfig("default")
 .setMergePolicy(
  LatestUpdateMapMergePolicy.class.getCanonicalName()
 );

Selected WAN replication configuration

<map name="default">
 <wan-replication-ref name="our-global-cluster">
  <merge-policy>
   com.hazelcast.map.merge.LatestUpdateMapMergePolicy
  </merge-policy>
 </wan-replication-ref>
</map>

WanReplicationRef wrr = new WanReplicationRef();
wrr.setName("our-global-cluster");
wrr.setMergePolicy(
 LatestUpdateMapMergePolicy.class.getCanonicalName()
);
config.getMapConfig("default")
 .setWanReplicationRef(wrr);

Indexes on values

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

config.getMapConfig("default")
 .addMapIndexConfig(new MapIndexConfig("name", false))
 .addMapIndexConfig(new MapIndexConfig("population", true));
Previous Section
End of Section
Next Section