🚀
0.4
Released on
- The cachelot has also been enabled to throw errors when out of memory or evict old items to free space for new ones. The initial version of search caching is enabled. This will address search memory overrun & will expand the query time window.
There are no bug fixes.
- Search config changes:
- name: ops.index.memory.maxmbcontrols the maximum size of the cache
- name: ops.index.memory.init.countcontrols number of index files warming done
- Recommendation for search runs
- Cache setting for 1GB search Pod: memory.maxmb – 256MB memory.init.count
- Cache setting for 2GB search Pod: memory.maxmb – 1024MB memory.init.count
Last modified 1yr ago