Almost done reallocating my drives to Ceph. Just two OSDs left...pic.twitter.com/BrhbJawIhX
-
mmm very dependent on workload i guess. high write high read with only one node in cluster was causing huge issues for us
Write IOPS/throughout went up by 2x with BlueStore, so that probably helps :-)
-
-
But it certainly has a performance overhead over raw storage, there's no getting around that. I get <50% the write throughout, haven't done any IOPS tests since that isn't too critical for me. I do find CephFS metadata perf to be excellent (I think it caches a lot).
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.