----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.csiden.org/r/118/#review290 -----------------------------------------------------------
Ship it! for HDDs, you'll find diminishing returns for zfs_vdev_async_read_max_active > 4, but the improvement is still useful - Richard Elling On Oct. 19, 2014, 6:20 a.m., Matthew Ahrens wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.csiden.org/r/118/ > ----------------------------------------------------------- > > (Updated Oct. 19, 2014, 6:20 a.m.) > > > Review request for OpenZFS Developer Mailing List and Christopher Siden. > > > Bugs: 5243 > https://www.illumos.org/projects/illumos-gate//issues/5243 > > > Repository: illumos-gate > > > Description > ------- > > 5243 zdb -b could be much faster > Reviewed by: Christopher Siden <christopher.si...@delphix.com> > Reviewed by: George Wilson <george.wil...@delphix.com> > > Original author: Matthew Ahrens > > > Diffs > ----- > > usr/src/uts/common/fs/zfs/dmu_traverse.c > fb1ec0755cf179b9e7dfc465d9f62a59a76fb408 > usr/src/cmd/zdb/zdb.c 63752a51ded0f319e56e0055bf2a50afb33641d7 > > Diff: https://reviews.csiden.org/r/118/diff/ > > > Testing > ------- > > ztest > zfs test suite > > ``` > on a pool with 5 million blocks: > old new improvement > zdb -b: 135sec 48sec 3x > zdb -bL: 93sec 16sec 6x > UMEM_DEBUG= zdb -b: 88sec 36sec 2x > UMEM_DEBUG= zdb -bL: 62sec 16sec 4x > ``` > > > Thanks, > > Matthew Ahrens > >
_______________________________________________ developer mailing list developer@open-zfs.org http://lists.open-zfs.org/mailman/listinfo/developer
