Hi Sage and Mark,
We had another test case (24 hours filestore run with journal stored in Optane drive) done, under our write intensive workloads, we saw IOPS increased by 59%, average latency reduced by 37%, but tail latency (99.99th) went up 65%. This new result is more meaningful for our production consideration.
Do you have any thoughts on why tail latency went up?
Thanks,
Jianjian
On 6/29/17, 9:20 AM, "LIU, Fei" <james.liu@xxxxxxxxxxxxxxx> wrote:
Hi All,
Below link is the slides we presented today in Ceph weekly performance meeting regarding to Ceph Filestore optimization with Intel drive and scrubbing optimization.
https://www.slideshare.net/jupiturliu/ceph-filestore-with-optane-drive-scrub-optimization
Please feel free to let us know if you have any questions.
Regards,
James
Alibaba Storage Team
--
To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
