IO500 Web Design Volunteer Requested
by committee＠io500.org
Dear IO500 Community,
We hope this email finds all of you well. We are proud to be a part of
this wonderful community and are thinking of all of you during this
challenging time. The call for submissions for the ISC20 list will be
sent out shortly in a separate email.
However, we wish to take this opportunity to discuss another matter. As
we presented during our SC19 BoF, we will be migrating our webpages from
the current VI4IO hosting site (thanks VI4IO!) to reside at their
permanent residence at io500.org (and maintain a mirror at VI4IO). To
progress towards this goal, we have created a Github repository for the
webpage code and a design document describing our vision for the webpage
(using PHP for the webpages with a SQL backend).
At this time, we are soliciting one or more volunteers in the community
to work with us to actually implement the new website following the
design document. If anyone is interested in volunteering for this role
or learning more about it, please contact us at <committee(a)io500.org>.
Stay safe and healthy all,
The IO500 Committee
2 years, 11 months
Separate "find" scores for easy/hard?
by Mark Nelson
Hi Guys,
I've been working on improving CephFS scores during the hard portions of
the test suite and noticed something interesting. We currently have a
fairly large discrepancy between easy and hard mdtest write tests
meaning we also have a fairly large disparity in the amount of data
written during each test. What I've noticed is that as I've improved
the throughput of mdtest hard writes, our find throughput has dropped.
You can see an example of that happening in the results posted here:
https://github.com/ceph/ceph/pull/34574
That makes sense if we are slower at finding files in a single giant
directory vs spread across many directories being serviced by dedicated
MDSes. IE I think the find score is probably overinflated when
significantly less data is written during the hard mdtest write phase
relative to the easy write phase. Does this seem like a reasonable
hypothesis? If so, it seems like this could be solved by running
separate find tests across the easy and hard datasets and scoring them
in the usual way.
Thoughts?
Mark
2 years, 11 months