Re: 8.3 / 8.2.6 restore comparison
От | Jeff Davis |
---|---|
Тема | Re: 8.3 / 8.2.6 restore comparison |
Дата | |
Msg-id | 1203977158.7878.59.camel@dogma.ljc.laika.com обсуждение исходный текст |
Ответ на | Re: 8.3 / 8.2.6 restore comparison ("Joshua D. Drake" <jd@commandprompt.com>) |
Ответы |
Re: 8.3 / 8.2.6 restore comparison
|
Список | pgsql-hackers |
On Mon, 2008-02-25 at 12:28 -0800, Joshua D. Drake wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > On Mon, 25 Feb 2008 12:17:10 -0800 > Jeff Davis <pgsql@j-davis.com> wrote: > > > > > I would personally rather keep it simple, hard core, and data > > > shoving as possible without any issue with scheduling etc.. > > > > > > > Just a thought. After it's actually implemented it won't be hard to > > see if it's a win. > > Yep :) but as a note: > > I am currently testing on the data set that is giving us all these > issues. Previously we were pushing ~ 22G an hour over a single thread. > I am currently pushing ~ 28G every 16 minutes over 6 threads. > > With 30-40% IO wait. That begs the question: what about 12 threads? Regards,Jeff Davis
В списке pgsql-hackers по дате отправления: