Chapter 18. Cache Store Implementations
sharedis set to
false), on node join, stale entries which might have been removed from the cluster might still be present in the stores and can reappear.
18.1. Cache Store Comparison
- The Single File Cache Store is a local file cache store. It persists data locally for each node of the clustered cache. The Single File Cache Store provides superior read and write performance, but keeps keys in memory which limits its use when persisting large data sets at each node. See Section 18.4, “Single File Cache Store” for details.
- The LevelDB file cache store is a local file cache store which provides high read and write performance. It does not have the limitation of Single File Cache Store of keeping keys in memory. See Section 18.5, “LevelDB Cache Store” for details.
- The JDBC cache store is a cache store that may be shared, if required. When using it, all nodes of a clustered cache persist to a single database or a local JDBC database for every node in the cluster. The shared cache store lacks the scalability and performance of a local cache store such as the LevelDB cache store, but it provides a single location for persisted data. The JDBC cache store persists entries as binary blobs, which are not readable outside JBoss Data Grid. See Section 18.6, “JDBC Based Cache Stores” for details.
- The JPA Cache Store (supported in Library mode only) is a shared cache store like JDBC cache store, but preserves schema information when persisting to the database. Therefore, the persisted entries can be read outside JBoss Data Grid. See Section 18.8, “JPA Cache Store” for details.