Re: Wait free LW_SHARED acquisition - v0.9
От | Andres Freund |
---|---|
Тема | Re: Wait free LW_SHARED acquisition - v0.9 |
Дата | |
Msg-id | 20141017181137.GE2075@alap3.anarazel.de обсуждение исходный текст |
Ответ на | Re: Wait free LW_SHARED acquisition - v0.9 (Amit Kapila <amit.kapila16@gmail.com>) |
Ответы |
Re: Wait free LW_SHARED acquisition - v0.9
|
Список | pgsql-hackers |
On 2014-10-17 17:14:16 +0530, Amit Kapila wrote: > On Tue, Oct 14, 2014 at 11:34 AM, Amit Kapila <amit.kapila16@gmail.com> > wrote: > > > > > > I am not sure why we are seeing difference even though running > > on same m/c with same configuration. > > I have tried many times, but I could not get the numbers you have > posted above with HEAD, however now trying with the latest version > [1] posted by you, everything seems to be fine at this workload. > The data at higher client count is as below: I'll try to reproduce it next week. But I don't think it matters all that much. Personally so far the performance numbers don't seem to indicate much reason to wait any further. We sure improve further, but I don't see much reason to wait because of that. > HEAD – commit 494affb > > Shared_buffers=8GB; Scale Factor = 3000 > > Client Count/No. Of Runs (tps) 64 128 Run-1 271799 247777 Run-2 274341 > 245207 Run-3 275019 252258 > > > > > > HEAD – commit 494affb + wait free lw_shared_v2 > > Shared_buffers=8GB; Scale Factor = 3000 > > Client Count/No. Of Runs (tps) 64 128 Run-1 286209 274922 Run-2 289101 > 274495 Run-3 289639 273633 So here the results with LW_SHARED were consistently better, right? You saw performance degradations here earlier? > So I am planning to proceed further with the review/test of your > latest patch. > According to me, below things are left from myside: > a. do some basic tpc-b tests with patch > b. re-review latest version posted by you Cool! > I know that you have posted optimization into StrategyGetBuffer() in > this thread, however I feel we can evaluate it separately unless you > are of opinion that both the patches should go together. > > [1] > http://www.postgresql.org/message-id/20141010111027.GC6670@alap3.anarazel.de No, I don't think they should go together - I wrote that patch because it was the bottleneck in the possibly regressing test and I wanted to see the full effect. Although I do think we should apply it ;) Greetings, Andres Freund -- Andres Freund http://www.2ndQuadrant.com/PostgreSQL Development, 24x7 Support, Training & Services
В списке pgsql-hackers по дате отправления: