Hitachi

In-Memory Data Grid Hitachi Elastic Application Data Store


10.9 Reducing the data usage of cache data files (performing compaction on cache data files)

This section is applicable when you will be using disk caches or two-way caches.

With disk caches and two-way caches, data is stored in cache data files by using an appending methodology. This means that when data is updated or deleted, invalid data remains in the files. Deleting such invalid data to reduce the data usage of the cache data files is called compaction.

The following figure illustrates compaction processing.

Figure 10‒2: Compaction processing

[Figure]

Note that the file size specified in the eads.cache.disk.filesize parameter in the cache properties remains unchanged after compaction.

Reference note

EADS employs an appending methodology to update cache data files. Data that is deleted using this method becomes invalid, but it is not actually deleted from the disk. When data is updated, the existing data that is updated becomes invalid and new data is added. The existing data that has become invalid is not deleted from the disk.

By performing compaction periodically, you can remove invalid data and use the disk efficiently.

Organization of this section