Ceph currently works with 3 types of backing store: FileStore (the default implementation which interacts with the filesystem, MemStore (mainly for performance testing) and the experimental key/value store (for use with RocksDB, LevelDB, Kinetic etc).
NewStore seeks to replace these all with a higher-performance general-purpose backend store which will mitigate the need for a journal (by using LevelDB for journalling). The speed improvements should benefit RGW, RBD and improve performance with SSDs/Flash etc
I made a few edits to the google doc. Otherwise looks ok!
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.
For information on the advisory, and where to find the updated
files, follow the link below.
If the solution does not work for you, open a new bug report.