Re: [REVIEW] Re: Compression of full-page-writes

Поиск
Список
Период
Сортировка
От Rahila Syed
Тема Re: [REVIEW] Re: Compression of full-page-writes
Дата
Msg-id CAH2L28v05JPycZ1qxUGP1C9t1EKvkfqtRo1=xvWHy1S=-fw3kA@mail.gmail.com
обсуждение исходный текст
Ответ на Re: [REVIEW] Re: Compression of full-page-writes  (Bruce Momjian <bruce@momjian.us>)
Ответы Re: [REVIEW] Re: Compression of full-page-writes  (Bruce Momjian <bruce@momjian.us>)
Список pgsql-hackers
Дерево обсуждения
>I am sorry but I can't understand the above results due to wrapping.
>Are you saying compression was twice as slow?

CPU usage at user level (in seconds)  for compression set 'on' is 562 secs
while that for compression  set 'off' is 354 secs. As per the readings, it takes little less than double CPU time to compress.
However , the total time  taken to run 250000 transactions for each of the scenario is as follows,

compression = 'on'  : 1838 secs
                    = 'off'  : 1701 secs                   

Different is around 140 secs.

Thank you,
Rahila Syed


On Wed, Dec 10, 2014 at 7:55 PM, Bruce Momjian <bruce@momjian.us> wrote:
On Wed, Dec 10, 2014 at 07:40:46PM +0530, Rahila Syed wrote:
> The tests ran for around 30 mins.Manual checkpoint was run before each test.
>
> Compression   WAL generated    %compression    Latency-avg   CPU usage
> (seconds)                                          TPS              Latency
> stddev               
>
>
> on                  1531.4 MB          ~35 %                  7.351 ms     
>   user diff: 562.67s     system diff: 41.40s              135.96            
>   13.759 ms
>
>
> off                  2373.1 MB                                     6.781 ms    
>       user diff: 354.20s      system diff: 39.67s            147.40            
>   14.152 ms
>
> The compression obtained is quite high close to 35 %.
> CPU usage at user level when compression is on is quite noticeably high as
> compared to that when compression is off. But gain in terms of reduction of WAL
> is also high.

I am sorry but I can't understand the above results due to wrapping.
Are you saying compression was twice as slow?

--
  Bruce Momjian  <bruce@momjian.us>        http://momjian.us
  EnterpriseDB                             http://enterprisedb.com

  + Everyone has their own god. +

В списке pgsql-hackers по дате отправления:

Предыдущее
От: Josh Berkus
Дата:
Сообщение: Re: Commitfest problems
Следующее
От: Heikki Linnakangas
Дата:
Сообщение: Re: WRITE_UINT_FIELD used where WRITE_OID_FIELD likely intended