Home > Storage > PowerScale (Isilon) > Product Documentation > Storage (general) > PowerScale OneFS SmartFlash > L3 cache persistence
Caching is predicated on keeping hot data hot. The most frequently accessed data and metadata on a node should remain in L2 cache and not get evicted to L3 cache. The next tier of cached data is accessed frequently enough to live in L3 cache, but not frequently enough to always live in RAM. A mechanism is in place to keep these semifrequently accessed blocks in L3 cache.
To maintain this L3 cache persistence, when the kernel goes to read a metadata or data block, the following steps are performed:
This marking process helps guard against the chronological eviction of blocks that are accessed while they are in the last 10% of the cache. It also serves to keep most of the useful data in cache.