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 CAH2L28v82CNxHD6+rnOEMfTZfEYKS+Ce-Nm6FWASixpDq3ouZQ@mail.gmail.com
обсуждение исходный текст
Ответ на Re: [REVIEW] Re: Compression of full-page-writes  (Alvaro Herrera <alvherre@2ndquadrant.com>)
Ответы Re: [REVIEW] Re: Compression of full-page-writes  ("Syed, Rahila" <Rahila.Syed@nttdata.com>)
Список pgsql-hackers
<p dir="ltr"><br /> Hello All,<p dir="ltr"> <br /> >Well, from Rahila's point of view the patch is actually
attached,but <br /> >she's posting from the Nabble interface, which mangles it and turns into <br /> >a link
instead.<pdir="ltr">Yes.<p dir="ltr"> <p dir="ltr">>but the end result is the <br /> >same: to properly submit a
patch,you need to send an email to the <br /> > mailing list, not join a group/forum from <br /> >some
intermediarynewsgroup site that mirrors the list.<br />  <p dir="ltr">Thank you. I will take care of it henceforth.  <p
dir="ltr">Pleasefind attached the patch to compress FPW.  Patch submitted by Fujii-san earlier in the thread is used to
mergecompression GUC with full_page_writes.<p dir="ltr"> <p dir="ltr">I am reposting the measurement numbers.<p
dir="ltr">ServerSpecification: <br /> Processors:Intel® Xeon ® Processor E5-2650 (2 GHz, 8C/16T, 20 MB) * 2 nos <br />
RAM:32GB <br /> Disk : HDWD     450GB 10K Hot Plug 2.5-inch SAS HDD * 8 nos <br /> 1 x 450 GB SAS HDD, 2.5-inch, 6Gb/s,
10,000rpm <p dir="ltr">Checkpoint segments: 1024 <br /> Checkpoint timeout: 5 mins <p dir="ltr">pgbench -c 64 -j 64 -r
-T900 -M prepared <br /> Scale factor: 1000 <p dir="ltr">                                WAL generated (MB)
          Throughput(tps)              Latency(ms) <br /> On                             9235.43                      
       979.03                                           65.36 <br /> Compress(pglz)       6518.68                      
        1072.34                                         59.66 <br /> Off                             501.04            
                    1135.17                                        56.34 <p dir="ltr">The results show  around 30
percentdecrease in WAL volume due to compression of FPW.<p dir="ltr">Thank you ,<p dir="ltr">Rahila Syed<p
dir="ltr">TomLane wrote:<br /> > Rahila Syed <<a href="mailto:rahilasyed.90@gmail.com">rahilasyed</a><a
href="mailto:rahilasyed.90@gmail.com">.90@</a><ahref="mailto:rahilasyed.90@gmail.com">gmail.com</a>> writes:<br />
>> Please find attached patch to compress FPW using pglz compression.<br /> ><br /> > Patch not actually
attachedAFAICS (no, a link is not good enough).<p dir="ltr">Well, from Rahila's point of view the patch is actually
attached,but<br /> she's posting from the Nabble interface, which mangles it and turns into<br /> a link instead.  Not
herfault, really -- but the end result is the<br /> same: to properly submit a patch, you need to send an email to
the<br/><a href="mailto:pgsql-hackers@postgresql.org">pgsql</a><a href="mailto:pgsql-hackers@postgresql.org">-</a><a
href="mailto:pgsql-hackers@postgresql.org">hackers</a><ahref="mailto:pgsql-hackers@postgresql.org">@</a><a
href="mailto:pgsql-hackers@postgresql.org">postgresql.org</a>mailinglist, not join a group/forum from<br /> some
intermediarynewsgroup site that mirrors the list.<p dir="ltr">--<br /> Álvaro Herrera               <a
href="http://www.2ndQuadrant.com/">http://www.2ndQuadrant.com/</a><br/> PostgreSQL Development, 24x7 Support, Training
&Services<br /><br /> 

В списке pgsql-hackers по дате отправления:

Предыдущее
От: Marko Tiikkaja
Дата:
Сообщение: Re: libpq connection status and closed fd
Следующее
От: "Syed, Rahila"
Дата:
Сообщение: Re: [REVIEW] Re: Compression of full-page-writes