Re: Move PinBuffer and UnpinBuffer to atomics

On Mon, Apr 25, 2016 at 6:04 PM, Alexander Korotkov <
a(dot)korotkov(at)postgrespro(dot)ru> wrote:

> On Sun, Apr 17, 2016 at 7:32 PM, Amit Kapila <amit(dot)kapila16(at)gmail(dot)com>
> wrote:
>
>> On Thu, Apr 14, 2016 at 8:05 AM, Andres Freund <andres(at)anarazel(dot)de>
>> wrote:
>> >
>> > On 2016-04-14 07:59:07 +0530, Amit Kapila wrote:
>> > > What you want to see by prewarming?
>> >
>> > Prewarming appears to greatly reduce the per-run variance on that
>> > machine, making it a lot easier to get meaningful results.
>> >
>>
>> I think you are referring the tests done by Robert on power-8 m/c, but
>> the performance results I have reported were on intel x86. In last two
>> days, I have spent quite some effort to do the performance testing of this
>> patch with pre-warming by using the same query [1] as used by Robert in his
>> tests. The tests are done such that first it start server, pre-warms the
>> relations, ran read-only test, stop server, again repeat this for next test.
>>
>
> What did you include into single run: test of single version (HEAD or
> Patch) or test of both of them?
>

single run includes single version (either HEAD or Patch).

>
>
>> I have observed that the variance in run-to-run performance still occurs
>> especially at higher client count (128). Below are results for 128 client
>> count both when the tests ran first with patch and then with HEAD and vice
>> versa.
>>
>> Test-1
>> ----------
>> client count - 128 (basically -c 128 -j 128)
>>
>> first tests ran with patch and then with HEAD
>>
>> Patch_ver/Runs HEAD (commit -70715e6a) Patch
>> Run-1 156748 174640
>> Run-2 151352 150115
>> Run-3 177940 165269
>>
>>
>> Test-2
>> ----------
>> client count - 128 (basically -c 128 -j 128)
>>
>> first tests ran with HEAD and then with patch
>>
>> Patch_ver/Runs HEAD (commit -70715e6a) Patch
>> Run-1 173063 151282
>> Run-2 173187 140676
>> Run-3 177046 166726
>>
>> I think this patch (padding pgxact) certainly is beneficial as reported
>> above thread. At very high client count some variation in performance is
>> observed with and without patch, but I feel in general it is a win.
>>
>
> So, what hardware did you use for these tests: power-8 or x86?
>

x86

> How long was single run?
>

5 minutes.

With Regards,
Amit Kapila.
EnterpriseDB: http://www.enterprisedb.com