Package net.sf.ehcache.distribution

This package is for cache replication.

See:
Description

Interface Summary

CacheManagerPeerListener	A listener for updates, which controls remote cache peers.
CacheManagerPeerProvider	Provides a discovery service to locate CachePeer listener peers for a Cache.
CachePeer	An interface for a cache peer to which updates are made remotely.
CacheReplicator	Replicates cache entries to peers of the CacheManager

Class Summary

CacheManagerPeerListenerFactory	An abstract factory for creating cache manager peer listeners.
CacheManagerPeerProviderFactory	An abstract factory for creating peers.
ConfigurableRMIClientSocketFactory	Default socket timeouts are unlikely to be suitable for cache replication.
EventMessage	An Event Message, in respect of a particular cache.
LegacyEventMessage	An Event Message, in respect of a particular cache.
ManualRMICacheManagerPeerProvider	A provider of Peer RMI addresses based off manual configuration.
MulticastKeepaliveHeartbeatReceiver	Receives heartbeats from any MulticastKeepaliveHeartbeatSenders out there.
MulticastKeepaliveHeartbeatSender	Sends heartbeats to a multicast group containing a compressed list of URLs.
MulticastRMICacheManagerPeerProvider	A peer provider which discovers peers using Multicast.
MulticastRMICacheManagerPeerProvider.CachePeerEntry	Entry containing a looked up CachePeer and date
RMIAsynchronousCacheReplicator	Listens to CacheManager and Cache events and propagates those to CachePeer peers of the Cache asynchronously.
RMIBootstrapCacheLoader	Loads Elements from a random Cache Peer
RMIBootstrapCacheLoaderFactory	A factory to create a configured RMIBootstrapCacheLoader
RMICacheManagerPeerListener	A cache server which exposes available cache operations remotely through RMI.
RMICacheManagerPeerListenerFactory	Builds a listener based on RMI.
RMICacheManagerPeerProvider	A provider of Peer RMI addresses.
RMICacheManagerPeerProviderFactory	Builds a factory based on RMI
RMICachePeer	An RMI based implementation of CachePeer.
RMICacheReplicatorFactory	Creates an RMICacheReplicator using properties.
RmiEventMessage
RMISynchronousCacheReplicator	Listens to CacheManager and Cache events and propagates those to CachePeer peers of the Cache.
TransactionalRMICachePeer	An RMI based implementation of CachePeer supporting transactions.

Enum Summary

RmiEventMessage.RmiEventType

Enumeration of event types.

Exception Summary

RemoteCacheException

A Cache Exception in the distribution mechanism.

Package net.sf.ehcache.distribution Description

This package is for cache replication.

Overview

Problems with Instance Caches in a Clustered Environment

Many production applications are deployed in clusters. If each application maintains its own cache, then updates made to one cache will not appear in the others. A workaround for web based applications is to use sticky sessions, so that a user, having established a session on one server, stays on that server for the rest of the session. A workaround for transaction processing systems using Hibernate is to do a session.refresh on each persistent object as part of the save. session.refresh explicitly reloads the object from the database, ignoring any cache values.

Replicated Cache

Another solution is to replicate data between the caches to keep them consistent. This is sometimes called cache coherency. Applicable operations include:

1. put
2. update (put which overwrites an existing entry)
3. remove

Replicated Cache Terms

Replicated Cache - a cache instance that notifies others when its contents change
Notification - a mechanism to replicate changes
Topology - a layout for how replicated caches connect with and notify each other

Notification Strategies

The best way of notifying of put and update depends on the nature of the cache.

If the Element is not available anywhere else then the Element itself should form the payload of the notification. An example is a cached web page. This notification strategy is called copy.

Where the cached data is available in a database, there are two choices. Copy as before, or invalidate the data. By invalidating the data, the application tied to the other cache instance will be forced to refresh its cache from the database, preserving cache coherency. Only the Element key needs to be passed over the network.

ehcache supports notification through copy and invalidate, selectable per cache.

Topology Choices

Peer Cache Replicator

Each replicated cache instance notifies every other cache instance when its contents change. This requires n-1 notifications per change, where n is the number of cache instances in the cluster.

Centralised Cache Replicator

Each replicated cache instance notifies a master cache instance when its contents change. The master cache then notifies the other instances. This requires n-1 notifications per change, where n is the number of cache instances in the cluster.

ehcache uses a peer replication topology. It adds a twist with CachePeerProvider, an interface which supplies a list of cache instance peers, so as to handle peers entering and leaving the cluster. Some ideas for peer provider implementations are: configuration time list, multicast discovery, application specific cluster list.

Replication Drawbacks and Solutions in ehcache's implementation

Some potentially significant obstacles have to be overcome if replication is to provide a net benefit.

Chatty Protocol

n-1 notifications need to happen each time a a cache instance change occurs. A very large amount of network traffic can be generated.

ehcache will buffer changes to lower chattiness.

Redundant Notifications

The cache instance that initiated the change should not receive its own notifications. To do so would add additional overhead. Also, notifications should not endlessly go back and forth as each cache listener gets changes caused by a remote replication.

ehcache CachePeerProvider indentifies the local cache instance and excludes it from the notification list. Each Cache has a GUID. That GUID can be compared with list of cache peers and the local peer excluded.

Infinite notifications are prevented by having each CacheReplicatorListener call putQuiet and removeQuite methods on their decorated caches, so as not to nofify listeners.

Potential for Inconsisent Data

Timing scenarios, race conditions, delivery and reliability constraints, and concurrent updates to the same cached data can cause inconsistency (and thus a lack of coherency) across the cache instancies. Acknowledgement: Much of the material here was drawn from Data Access Patterns, by Clifton Nock.