Multiple fixes around remote execution, remote listeners and iteration.

Improved failover handling logic to correctly deal with socket error and remote disconnect situations.

Removed getVersioned, getBulk and getBulkKeys methods since these are deprecated in favour of getWithMetadata and getAll. The examples in the README file have been updated.

Added API documentation for Javascript client. Online API docs for Javascript client can be found here.

SNI support for HotRod, which paves the way for server multi-tenancy

JDBC cache stores can use upsert (one of our oldest bugs !!!)

An httpd-style access log for server

The new sequential interceptor architecture

Lucene query caching

A grand total of 138 issues were resolved

Graceful clustered shutdown / restart with persistent state

The REST cache store has a shiny new Netty-based backend, courtesy of Antoine Toulme

and many, many bugfixes: consult our issue tracker for a list of things we have fixed. 

We’ve removed Spring 3.x support

We’ve removed our home-brew map/reduce implementation. By now you should be using the wonderful Streams-based API instead which are much nicer and provide better performance

We’ve dropped our custom NotifyingListeners in favour of Java’s standard CompletableFuture

Fixed cluster startup in client server mode

Fixed default cache race condition during server start

Fixed QueryDSL parser failure with inheritence

Improved test suite stability

Fixed registering clustered listener (previously it cased SuspectException)

Protostream was upgraded to 3.0.5.Final

XStream was upgraded to 1.4.9

Updated README file containing simpler individual examples on each main feature set.

Remote execution - Users can now load remote scripts into Infinispan Servers using the addScript method and these scripts can be remotely executed via the execute operation. An example of this can be found in the README file.

Server-side statistic can now be retrieved using the stats operation. Again, an example can be found in the README file.

The client now supports cluster topology updates, so it can handle remote notifications of server side topology changes, including addition or removal of server-side cluster nodes.

Using the cluster topology information and hashing algorithms, the client can now apply the same Consistent-Hash algorithm servers use to distribute data around the cluster, and hence can immediately locate the server where a particular key lives.

New algorithm for remote reads in distributed mode that reduces the network usage

From now on, the consistent hash is shared among caches with the same membership - entries in different caches using the same key will be co-located

A new lambda-enabled Cluster Executor which does not require a clustered cache, but can be constructed directly from a DefaultCacheManager.

Protostream 3.0.5, Lucene 5.5, Hibernate Search 5.6.0.Alpha2

And many tweaks and improvements overall.

Persistent login session which survives browser refresh

Deployment / undeployment of user code

Management of ProtoBuf schemas for indexing

Management of Scripts for task execution

Event viewing

Much improved cache and template configuration

Cache container configuration

PatternFly 3.2

Full CRUD operation support, e.g. put, get, remove, containsKey…​etc.

Compare-And-Swap operation support, e.g. putIfAbsent, getWithVersion/getWithMetadata, replace, replaceWithVersion, removeWithVersion…​etc.

Expiration with absolute lifespan or relative maximum idle time is supported. This expiration parameters as passed as optional parameters to create/update methods and they support multiple time units, e.g. \{lifespan: '1m', maxIdle: '1d'}.

Update and remove operations can optionally return previous values by passing in\{previous: true} option.

Bulk store/retrieve/delete operations are supported, e.g. putAll, getAll, getBulk, getBulkKeys, clear…​etc.

Cache contents can be iterated over using the iterator method.

Cache size can be determined using the size method.

Remote cache listeners can be plugged using the addListener method, which takes the event type (create, modify, remove or expiry) and the function callback as parameter.

lots of enhancements and additions to the administration console for Infinispan Server, such as cluster event views, task management, container administration and deployments. Some of these will be further refined before the upcoming Final release. If you prefer to manage your servers from the good ol' command-line, the CLI commands have also been similarly enhanced.

performance improvements for clustered reads, bulk and stream ops, and a sensible reduction in memory allocation rate. More improvements will come as we are rewriting some parts of the core to allow further optimizations.

a new distributed executor, ClusterExecutor which does not require an existing distributed cache and which comes complete with Java 8 goodness such as CompletableFutures and lambda support. This nicely complements our existing Streams support.

a new, experimental index manager to store indexes on Infinispan with better scalability and indexing throughput. Additionally we’ve upgraded Lucene to 5.4.1 and Hibernate Search to 5.6.0.Alpha1

the ability to indicate a number of nodes for initial cluster formation

lots of bug fixes

SyncConsistentHashFactory is now the default consistent hash factory (ISPN-4851). This means the owners of a key are going to be the same in all the caches, as long as the caches have the same members and the same number of owners.

We now have a new way for applications to determine the location of keys: the KeyPartitioner interface (ISPN-5465). In spirit it is very similar to the existing Grouper interface, but it is segment-aware. Note that none of the HotRod clients support Keypartitioner in this release.

We improved the rolling upgrade performance (ISPN-6046).

We added a new event logger that records cluster membership, cache topology, and cache availability changes (ISPN-5816). In the next release, it will also be accessible through the server’s management console.