From: | Fabien COELHO <coelho(at)cri(dot)ensmp(dot)fr> |
---|---|
To: | Aidan Van Dyk <aidan(at)highrise(dot)ca> |
Cc: | Andres Freund <andres(at)2ndquadrant(dot)com>, Josh Berkus <josh(at)agliodbs(dot)com>, PostgreSQL Developers <pgsql-hackers(at)postgresql(dot)org> |
Subject: | Re: postgresql latency & bgwriter not doing its job |
Date: | 2014-08-27 14:50:24 |
Message-ID: | alpine.DEB.2.10.1408271634540.8876@sto |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
Hello,
> If all you want is to avoid the write storms when fsyncs start happening on
> slow storage, can you not just adjust the kernel vm.dirty* tunables to
> start making the kernel write out dirty buffers much sooner instead of
> letting them accumulate until fsyncs force them out all at once?
I can try, when I have finished with the current round of testing.
Note that, as Andres put it, it currently "sucks". Having to tinker with
linux kernel parameters just to handle a small load without being offline
10% of the time does not look very good, so even if it works, ISTM that a
pg side solution is desirable.
--
Fabien.
From | Date | Subject | |
---|---|---|---|
Next Message | Robert Haas | 2014-08-27 14:52:28 | Re: [REVIEW] Re: Compression of full-page-writes |
Previous Message | Bruce Momjian | 2014-08-27 14:40:53 | Re: [BUGS] Re: BUG #9555: pg_dump for tables with inheritance recreates the table with the wrong order of columns |