Hello Rubio,
Thank you for your email. We look forward to your submission and will
conduct our analysis of the "shift" effects once all SC19 submissions have
been received.
Thanks,
The IO500 Committee
On Wed, Oct 30, 2019 at 10:03 PM Ruibo Wang via IO-500 <io-500(a)vi4io.org>
wrote:
Dear All,
I've built a cluster to test the IO500 benchmark.
The preliminary results showed that the new "shift" parameter did
influence the stat result badly.
If I use "-N" parameter, the mdtest result is:
mdtest_easy_stat 1429.930 kiops : time 253.86 seconds
mdtest_hard_stat 962.500 kiops : time 191.10 seconds
If I don't use "-N" parameter, the mdtest result is:
mdtest_easy_stat 200250.000 kiops : time 1.84 seconds
mdtest_hard_stat 51038.200 kiops : time 4.47 seconds
It is clear that the "shift" parameter change will impact the stat
result with the same hardware and software configuration. Now I'm still
trying to optimize the test. Hope I'll deliver a better result later.
Ruibo Wang
_______________________________________________
IO-500 mailing list
IO-500(a)vi4io.org
https://www.vi4io.org/mailman/listinfo/io-500