In a Flash Accel-enabled system, applications write to cache as well as to back-end storage, so that future application reads can be serviced locally from cache. For frequently accessed data, the result is increased throughput and reduced latency.
On an application read request, Flash Accel looks for the data in cache. If the data is found in cache, called a
cache hit, it is returned to the application. If the data is not found in cache, called a
cache miss, the data is retrieved from the storage system, returned to the application, and then inserted in the cache so it is available for future read requests.
On an application write request, Flash Accel looks for the data in cache. If the data is found in cache, it is invalidated in the cache. The new data is written to the storage system and then inserted in the cache so it is available for future read requests. If the data is not found in cache, it is written to the storage system and then inserted in cache.