Dear All, We have a small setup of lustre with 7 OSTs on 8gb FC . We have kept one OST per FC port. We have lustre 2.3 with CentOS 6.3. There are 32 clients which access this over FDR IB. We can achieve more than 1.3GB/s throughput using IOR, without cache. Which is roughly 185MB/s per OST. We wanted to know if this is normal. Should we expect more from 8gb FC port. OSTs are on 8+2 RAID6 . Thanks and Regards Dilip G Sathaye -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.lustre.org/pipermail/lustre-discuss/attachments/20130527/ef2e64c2/attachment.html
What is the raw performance of your pure storage system? Next step - run obdfilter-survey on your installation - this will be your storage performance baseline under lustre. Next step - run IOR on lustre clients again. Compare those 3 numbers. You will not achieve more from lustre clients then from storage performance if lustre servers are fast enough. P On Mon, May 27, 2013 at 7:36 AM, Dilip Sathaye <dsathaye-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:> Dear All, > > We have a small setup of lustre with 7 OSTs on 8gb FC . We have kept one OST > per FC port. We have lustre 2.3 with CentOS 6.3. There are 32 clients which > access this over FDR IB. We can achieve more than 1.3GB/s throughput using > IOR, without cache. Which is roughly 185MB/s per OST. We wanted to know if > this is normal. Should we expect more from 8gb FC port. OSTs are on 8+2 > RAID6 .
Seemingly Similar Threads
- Enable async journals
- Depreciated client still shown on OST exports
- [PATCH 0/1] drm/nouveau/disp: prefer identity-mapped route of SOR <-> macro link
- [PATCH 1/2] drm/nouveau/nvkm/outp: Use WARN_ON() in conditionals in nvkm_outp_init_route()
- 1.8.4 and write-through cache