Zheng, I don't have any idea what pieces have changed in that kernel range. Did we have to flip some switches that slowed things down and we expect to flip back, or did something more fundamental happen? Do these results make any sense? I'm a little surprised to find ceph-fuse that much faster than either kernel, but I've not checked the clock times in our own tests. -Greg On Fri, Jun 26, 2015 at 3:20 PM, Dan van der Ster <dan@xxxxxxxxxxxxxx> wrote: > Hi all, > > Today we are running some tests with this POSIX compatibility test [1] > and noticed a possible performance regression in the latest kernel. > The cluster we are testing is 0.94.2. > > 0.94.2 ceph-fuse client: > All tests successful. > Files=184, Tests=1957, 83 wallclock secs ( 0.72 usr 0.16 sys + 5.55 > cusr 10.17 csys = 16.60 CPU) > Result: PASS > > Kernel client 3.10.0-229.7.2.el7.x86_64: > All tests successful. > Files=184, Tests=1957, 143 wallclock secs ( 0.88 usr 0.15 sys + 5.86 > cusr 10.43 csys = 17.32 CPU) > Result: PASS > > Kernel client 4.1.0-1.el7.elrepo.x86_64: > All tests successful. > Files=184, Tests=1957, 378 wallclock secs ( 0.81 usr 0.34 sys + 5.30 > cusr 9.75 csys = 16.20 CPU) > Result: PASS > > Any idea why the kernel client is getting much slower? > > Cheers, Dan > > [1] http://tuxera.com/sw/qa/pjd-fstest-20080816.tgz > _______________________________________________ > ceph-users mailing list > ceph-users@xxxxxxxxxxxxxx > http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com _______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
 |