On Wed, Aug 24, 2016 at 2:24 PM, Heikki Linnakangas <hlinnaka@iki.fi> wrote:
>
> And here are the results on the 72 core machine (thanks again, Alexander!).
> The test setup was the same as on the 32-core machine, except that I ran it
> with more clients since the system has more CPU cores. In summary, in the
> best case, the patch increases throughput by about 10%. That peak is with 64
> clients. Interestingly, as the number of clients increases further, the gain
> evaporates, and the CSN version actually performs worse than unpatched
> master. I don't know why that is. One theory that by eliminating one
> bottleneck, we're now hitting another bottleneck which doesn't degrade as
> gracefully when there's contention.
>
Quite possible and I think it could be either due CLOGControlLock or
WALWriteLock. Are you planning to work on this for this release, if
not do you think meanwhile we should pursue "caching the snapshot"
idea of Andres? Last time Mithun did some benchmarks [1] after fixing
some issues in the patch and he found noticeable performance increase.
[1] - https://www.postgresql.org/message-id/CAD__Ouic1Tvnwqm6Wf6j7Cz1Kk1DQgmy0isC7%3DOgX%2B3JtfGk9g%40mail.gmail.com
--
With Regards,
Amit Kapila.
EnterpriseDB: http://www.enterprisedb.com