Whamcloud - gitweb
LU-4865 zfs: grow block size by write pattern 27/15127/12
authorJinshan Xiong <jinshan.xiong@intel.com>
Wed, 3 Jun 2015 07:06:19 +0000 (00:06 -0700)
committerOleg Drokin <oleg.drokin@intel.com>
Thu, 10 Sep 2015 01:02:45 +0000 (01:02 +0000)
commit3e4369135127b350dbc26a4a5dc94cfa46e394cf
tree540faa049f596faf8d16a932141a918dbd4df1d6
parent01ca899324738343279c1d63823b7fab937197dc
LU-4865 zfs: grow block size by write pattern

This patch grows the block size by write RPC. The osd-zfs blocksize
used to be fixed at 128KB, which is too big for random write and
too small for seqential write.

This patch decides the block size by the first few RPCs. If the first
few RPCs are sequential, mostly it will use bigger block size;
otherwise, smaller block size will be used.

Signed-off-by: Jinshan Xiong <jinshan.xiong@intel.com>
Change-Id: I6cf8fa8eca998e73bbb99f66eb1323c3abcf39f0
Reviewed-on: http://review.whamcloud.com/15127
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Alex Zhuravlev <alexey.zhuravlev@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
lustre/osd-zfs/osd_io.c
lustre/osd-zfs/osd_object.c