Re: pg_prewarm
От | Cédric Villemain |
---|---|
Тема | Re: pg_prewarm |
Дата | |
Msg-id | 201206202302.24009.cedric@2ndquadrant.com обсуждение исходный текст |
Ответ на | Re: pg_prewarm (Robert Haas <robertmhaas@gmail.com>) |
Список | pgsql-hackers |
<p style=" margin-top:0px; margin-bottom:0px; margin-left:0px; margin-right:0px; -qt-block-indent:0; text-indent:0px; -qt-user-state:0;">>The biggest problem with pgfincore from my point of view is that it<p style=" margin-top:0px; margin-bottom:0px;margin-left:0px; margin-right:0px; -qt-block-indent:0; text-indent:0px; -qt-user-state:0;">> only worksunder Linux, whereas I use a MacOS X machine for my<p style=" margin-top:0px; margin-bottom:0px; margin-left:0px; margin-right:0px;-qt-block-indent:0; text-indent:0px; -qt-user-state:0;">> development, and there is also Windows to thinkabout. Even if that<p style=" margin-top:0px; margin-bottom:0px; margin-left:0px; margin-right:0px; -qt-block-indent:0;text-indent:0px; -qt-user-state:0;">> were fixed, though, I feel we ought to have something in thecore<p style=" margin-top:0px; margin-bottom:0px; margin-left:0px; margin-right:0px; -qt-block-indent:0; text-indent:0px;-qt-user-state:0;">> distribution. This patch got more +1s than 95% of what gets proposed<p style=" margin-top:0px;margin-bottom:0px; margin-left:0px; margin-right:0px; -qt-block-indent:0; text-indent:0px; -qt-user-state:0;">>on hackers.<p style="-qt-paragraph-type:empty; margin-top:0px; margin-bottom:0px; margin-left:0px;margin-right:0px; -qt-block-indent:0; text-indent:0px; "> <p style=" margin-top:0px; margin-bottom:0px; margin-left:0px;margin-right:0px; -qt-block-indent:0; text-indent:0px; -qt-user-state:0;">pgfincore works also on BSD kernels.Can you try on your MacOSX ? (I don't have one here).<p style=" margin-top:0px; margin-bottom:0px; margin-left:0px;margin-right:0px; -qt-block-indent:0; text-indent:0px; -qt-user-state:0;">As of freeBSD 8.3 there is suportfor posix_fadvise call so both PostgreSQL core and pgfincore now support the preloading on this distribution (I'venot tested it recently but it should).<p style="-qt-paragraph-type:empty; margin-top:0px; margin-bottom:0px; margin-left:0px;margin-right:0px; -qt-block-indent:0; text-indent:0px; "> <p style=" margin-top:0px; margin-bottom:0px; margin-left:0px;margin-right:0px; -qt-block-indent:0; text-indent:0px; -qt-user-state:0;">All pgfincore features should nowworks in most places, except windows.<p style="-qt-paragraph-type:empty; margin-top:0px; margin-bottom:0px; margin-left:0px;margin-right:0px; -qt-block-indent:0; text-indent:0px; "> <p style=" margin-top:0px; margin-bottom:0px; margin-left:0px;margin-right:0px; -qt-block-indent:0; text-indent:0px; -qt-user-state:0;">> > Also, Robert has accuratelydescribed this as "mechanism, not policy".<p style=" margin-top:0px; margin-bottom:0px; margin-left:0px; margin-right:0px;-qt-block-indent:0; text-indent:0px; -qt-user-state:0;">> > That's fine, that's what all of our stuffis. Replication and most of<p style=" margin-top:0px; margin-bottom:0px; margin-left:0px; margin-right:0px; -qt-block-indent:0;text-indent:0px; -qt-user-state:0;">> > postgresql.conf suffers from that. Eventually someone willwant to<p style=" margin-top:0px; margin-bottom:0px; margin-left:0px; margin-right:0px; -qt-block-indent:0; text-indent:0px;-qt-user-state:0;">> > create a way to save and restore various caches across server restarts,<p style="margin-top:0px; margin-bottom:0px; margin-left:0px; margin-right:0px; -qt-block-indent:0; text-indent:0px; -qt-user-state:0;">>> as discussed before. Would that mean there will be a third way to do<p style=" margin-top:0px;margin-bottom:0px; margin-left:0px; margin-right:0px; -qt-block-indent:0; text-indent:0px; -qt-user-state:0;">>> all this, or could we at least align things a bit so that such a<p style=" margin-top:0px; margin-bottom:0px;margin-left:0px; margin-right:0px; -qt-block-indent:0; text-indent:0px; -qt-user-state:0;">> > facilitycould use most of the proposed prewarming stuff? (Patches for<p style=" margin-top:0px; margin-bottom:0px; margin-left:0px;margin-right:0px; -qt-block-indent:0; text-indent:0px; -qt-user-state:0;">> > the cache restoring exist,so it should be possible to predict this a<p style=" margin-top:0px; margin-bottom:0px; margin-left:0px; margin-right:0px;-qt-block-indent:0; text-indent:0px; -qt-user-state:0;">> > little bit.)<p style=" margin-top:0px;margin-bottom:0px; margin-left:0px; margin-right:0px; -qt-block-indent:0; text-indent:0px; -qt-user-state:0;">><p style=" margin-top:0px; margin-bottom:0px; margin-left:0px; margin-right:0px; -qt-block-indent:0;text-indent:0px; -qt-user-state:0;">> Well, pg_buffercache + pg_prewarm is enough to save and restoreshared<p style=" margin-top:0px; margin-bottom:0px; margin-left:0px; margin-right:0px; -qt-block-indent:0; text-indent:0px;-qt-user-state:0;">> buffers. Not the OS cache, but we don't have portable code to query<p style=" margin-top:0px;margin-bottom:0px; margin-left:0px; margin-right:0px; -qt-block-indent:0; text-indent:0px; -qt-user-state:0;">>the OS cache yet anyway.<p style="-qt-paragraph-type:empty; margin-top:0px; margin-bottom:0px; margin-left:0px;margin-right:0px; -qt-block-indent:0; text-indent:0px; "> <p style=" margin-top:0px; margin-bottom:0px; margin-left:0px;margin-right:0px; -qt-block-indent:0; text-indent:0px; -qt-user-state:0;">+pgfincore and the OS cache partis done.<p style="-qt-paragraph-type:empty; margin-top:0px; margin-bottom:0px; margin-left:0px; margin-right:0px; -qt-block-indent:0;text-indent:0px; "> <p style=" margin-top:0px; margin-bottom:0px; margin-left:0px; margin-right:0px; -qt-block-indent:0;text-indent:0px; -qt-user-state:0;">-- <p style=" margin-top:0px; margin-bottom:0px; margin-left:0px;margin-right:0px; -qt-block-indent:0; text-indent:0px; -qt-user-state:0;">Cédric Villemain +33 (0)6 20 3022 52<p style=" margin-top:0px; margin-bottom:0px; margin-left:0px; margin-right:0px; -qt-block-indent:0; text-indent:0px;-qt-user-state:0;">http://2ndQuadrant.fr/<p style=" margin-top:0px; margin-bottom:0px; margin-left:0px; margin-right:0px;-qt-block-indent:0; text-indent:0px; -qt-user-state:0;">PostgreSQL: Support 24x7 - Développement, Expertiseet Formation<p style="-qt-paragraph-type:empty; margin-top:0px; margin-bottom:0px; margin-left:0px; margin-right:0px;-qt-block-indent:0; text-indent:0px; ">
В списке pgsql-hackers по дате отправления:
Следующее
От: Tom LaneДата:
Сообщение: Remove support in ri_triggers.c for zero-column foreign keys?