On Mon, Apr 2, 2018 at 11:18 AM Robert Stanford <rstanford8...@gmail.com> wrote:
> > This is a known issue as far as I can tell, I've read about it several > times. Ceph performs great (using radosgw), but as the OSDs fill up > performance falls sharply. I am down to half of empty performance with > about 50% disk usage. > > My questions are: does adding more OSDs / disks to the cluster restore > performance? (Is it the absolute number of objects that degrades > performance, or % occupancy?) And, will the performance level off at some > point, or will it continue to get lower and lower as our disks fill? > There are two dimensions of interest for you here: 1) the speed of your bucket index, 2) the speed of the OSD store for reading/writing actual data. (2) should be better with bluestore than filestore AFAIK, and will definitely improve as you add more OSDs to the cluster. (1) will depend more on the amount of bucket index sharding and the number of buckets you're working with. -Greg > > Thank you > _______________________________________________ > ceph-users mailing list > ceph-users@lists.ceph.com > http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com >
_______________________________________________ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
