Ehcache is the most widely-used Java-based cache. It is:

Ehcache scales from in-process caching, all the way to mixed in-process/out-of-process deployments with terabyte-sized caches.

Ehcache's concurrency features are designed for large, high concurrency systems.

Many users of Ehcache hardly know they are using it. Sensible defaults require no initial configuration.

Ehcache strives to maintain a small footprint - keeping your apps as light as they can be.

The only dependency for core use is SLF4J.

The largest Ehcache installations utilize multiple terabytes of data storage.

With off-heap storage, Ehcache has been tested to store 6TB of data in a single process (JVM).

The largest Ehcache installations use hundreds of caches.

Ehcache is specifically built and tested to run well under highly concurrent access on systems with dozens of cores. This results in an optimal balance between thread safety and performance.

Provides multiple strategies for:

You can use Ehcache as a JCache provider. This allows you to use JCache API calls to develop a complete application, without the need to use any Ehcache API calls.

Ehcache supports simple yet high performance distributed caching.

High quality implementations for common caching scenarios and patterns.

This is the trusty old command pattern with a twist: asynchronous behavior, fault tolerance and caching. Creates a command, caches it and then attempts to execute it.

Ehcache is popularly used as Hibernate's second-level cache.

Ehcache supports JTA and is a fully XA compliant resource participating in the transaction, two-phase commit and recovery.

See the complete Transaction Module Java Documentation at www.ehcache.org.

The Javadoc documentation of the API can be found here:

There is an open-source version of Ehcache, paired with Terracotta Server open source functionality. This can be found at http://www.terracotta.org/open-source/.