On Sat, Oct 11, 2014 at 7:02 PM, Amit Kapila <amit.kapil...@gmail.com> wrote: > On Sat, Oct 11, 2014 at 6:40 PM, Andres Freund <and...@2ndquadrant.com> wrote: > > On 2014-10-11 07:26:57 +0530, Amit Kapila wrote: > > > On Sat, Oct 11, 2014 at 7:00 AM, Andres Freund <and...@2ndquadrant.com > > > > > And since > > > > your general performance numbers are a fair bit lower than what I see > > > > with, hopefully, the same code on the same machine... > > > > > > You have reported numbers at 1000 scale factor and mine were > > > at 3000 scale factor, so I think the difference is expected. > > > > The numbers for 3000 show pretty much the same: > > > > SCALE 128 160 175 > > HEAD 352113 339005 336491 > > LW_SHARED 365874 347931 342528 > > > > Hm. I wonder if you're using pgbench without -M prepared? > > No, I use below statement: > ./pgbench -c 128 -j 128 -T 300 -S -M prepared postgres > > > That'd about > > explain the difference. > > Here I think first thing to clarify is why the numbers on HEAD are > different?
I have taken the latest code and recreated the database and tried again on power-7 m/c (hydra) and below is the result: Result with -M prepared: Duration of each individual run - 5 mins HEAD – commit 494affb Shared_buffers=8GB; Scale Factor = 3000 Client Count/No. Of Runs (tps) 128 160 Run-1 258385 239908 Run-2 257835 238624 Run-3 255967 237905 Result without -M prepared: Duration of each individual run - 5 mins HEAD – commit 494affb Shared_buffers=8GB; Scale Factor = 3000 Client Count/No. Of Runs (tps) 128 160 Run-1 228747 220961 Run-2 229817 214464 Run-3 227386 216619 I am not sure why we are seeing difference even though running on same m/c with same configuration. I think there is some difference in the way we are running tests, if you don't mind could you please share the exact steps and non-default postgresql.conf settings with me. The below list of things could be useful for me to reproduce the numbers you are seeing: a. build steps (any script you are using) b. non-default postgresql.conf settings c. Exact pgbench statements used With Regards, Amit Kapila. EnterpriseDB: http://www.enterprisedb.com