Received: from makus.postgresql.org ([98.129.198.125]) by malur.postgresql.org with esmtp (Exim 4.72) (envelope-from ) id 1TUHwu-0001lD-LU for pgsql-performance@postgresql.org; Fri, 02 Nov 2012 14:10:16 +0000 Received: from mail-vb0-f46.google.com ([209.85.212.46]) by makus.postgresql.org with esmtp (Exim 4.72) (envelope-from ) id 1TUHws-0005Vs-Ms for pgsql-performance@postgresql.org; Fri, 02 Nov 2012 14:10:15 +0000 Received: by mail-vb0-f46.google.com with SMTP id ff1so3806940vbb.19 for ; Fri, 02 Nov 2012 07:10:12 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=praus.net; s=google; h=mime-version:x-originating-ip:in-reply-to:references:from:date :message-id:subject:to:cc:content-type; bh=IbDcFETNtQF+5Fu2e0wP0ybVJuGzhGkrvSp0Dr4Qv+s=; b=WUdeZ3po0f09ufY7L/GOtzTpkHVaCUGGYUZSDByRU6kmPhgcP1kuSoTv4HbWogpjrA VfJmhbWW8ry42r7Ai2nBd0ZaMGwwjyHzSyFGrja9B/eSjTBOZcQ5V+U0+KplLaSHIufV lQSoA1awe+HitTzy5zEufjqz3BKemVSvPjULI= X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:x-originating-ip:in-reply-to:references:from:date :message-id:subject:to:cc:content-type:x-gm-message-state; bh=IbDcFETNtQF+5Fu2e0wP0ybVJuGzhGkrvSp0Dr4Qv+s=; b=pFnxdmlqX5HmdQVKDQbvlWFF7G9FawRcYdRUzCGG8BTqqnRnkGzKHY1OWa/G1DWMLl JVsOaS3JZ5sqvhNe1OFhreB1R6/2DMZIrx+UNPuGOKJsmqk20OObaM59WptDzbyRFVgR ueiF0BM2I4YOWuwKwzJ9w+jhUtm6qThc4mcjaOFKwP4ciiXZmCg9a5KgzCTUFAv3OVcG Z4JQLo3UuqRfUke3z0aXkjMCCmEK1CfMkfNXp4reW99jXRizcT3SRx8tF3/SmXL46Y2I Quo8BPQ8/5heJIfduQqB6bfP6jcrZIuDnl73RT9nKHpJvwMif9RNlaMtvRyhPFi1Igeq 56kw== Received: by 10.52.92.11 with SMTP id ci11mr1695816vdb.7.1351865412537; Fri, 02 Nov 2012 07:10:12 -0700 (PDT) MIME-Version: 1.0 Received: by 10.58.143.14 with HTTP; Fri, 2 Nov 2012 07:09:52 -0700 (PDT) X-Originating-IP: [129.62.151.28] In-Reply-To: References: From: Petr Praus Date: Fri, 2 Nov 2012 09:09:52 -0500 Message-ID: Subject: Re: Increasing work_mem and shared_buffers on Postgres 9.2 significantly slows down queries To: Scott Marlowe Cc: pgsql-performance@postgresql.org Content-Type: multipart/alternative; boundary=20cf3071ce0ab2ce3204cd83b13a X-Gm-Message-State: ALoCoQlbqxCCDAf7rDWfT4hc5qJFETarjX0/x4dKd/u1KnHWMA/rR9DA7twVLCf5FgfJ4V155oJL X-Pg-Spam-Score: -2.6 (--) X-Archive-Number: 201211/28 X-Sequence-Number: 48394 --20cf3071ce0ab2ce3204cd83b13a Content-Type: text/plain; charset=UTF-8 I did run each query several times, the results I posted are for ~10th run of the query. The zone reclaim mode is 0. On 2 November 2012 00:39, Scott Marlowe wrote: > Two possibilities: > > caching. make sure to run each query several times in a row. > > zone reclaim mode. If this has gotten turned on turn it back off. > > How to tell: > > sysctl -n vm.zone_reclaim_mode > > Output should be 0. If it's not, then add this to /etc/sysctl.conf: > > vm.zone_reclaim_mode=0 > > and run: sudo sysctl -p > > and see if that helps. > --20cf3071ce0ab2ce3204cd83b13a Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable I did run each query several times, the results I posted are for ~10th run = of the query.

The zone reclaim mode is 0.


On 2 November 2012 00:= 39, Scott Marlowe <scott.marlowe@gmail.com> wrote:
Two possibilities:

caching. =C2=A0make sure to run each query several times in a row.

zone reclaim mode. If this has gotten turned on turn it back off.

How to tell:

sysctl -n vm.zone_reclaim_mode

Output should be 0. =C2=A0If it's not, then add this to /etc/sysctl.con= f:

vm.zone_reclaim_mode=3D0

and run: sudo sysctl -p

and see if that helps.

--20cf3071ce0ab2ce3204cd83b13a--