Received: from makus.postgresql.org (makus.postgresql.org [98.129.198.125]) by mail.postgresql.org (Postfix) with ESMTP id 0EE786B258B for ; Thu, 29 Dec 2011 12:05:54 -0400 (AST) Received: from s1v2.e-lietuva.lt ([82.135.156.154] helo=centos5.e-lietuva.lt) by makus.postgresql.org with esmtp (Exim 4.72) (envelope-from ) id 1RgIUK-0005Xr-Cm for pgsql-performance@postgresql.org; Thu, 29 Dec 2011 16:05:54 +0000 Received: (qmail 1585 invoked from network); 29 Dec 2011 18:05:37 +0200 Received: from ip-195-14-173-2.bnk.lt (HELO ?192.168.10.129?) (195.14.173.2) by s1v2.e-lietuva.lt with (DHE-RSA-AES256-SHA encrypted) SMTP; 29 Dec 2011 18:05:37 +0200 Message-ID: <4EFC8FD0.9020100@ruptela.lt> Date: Thu, 29 Dec 2011 18:05:36 +0200 From: Aleksej Trofimov User-Agent: Mozilla/5.0 (X11; Linux i686; rv:8.0) Gecko/20111124 Thunderbird/8.0 MIME-Version: 1.0 To: Claudio Freire CC: sgupta , pgsql-performance@postgresql.org Subject: Re: Postgresql Replication Performance References: <1325151184846-5107278.post@n5.nabble.com> <4EFC7A49.6070107@ruptela.lt> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Pg-Spam-Score: -1.9 (-) X-Archive-Number: 201112/202 X-Sequence-Number: 46001 On 12/29/2011 05:00 PM, Claudio Freire wrote: > Second, is that WAL activity on streaming replication or WAL shipping > is documented to contain more data than on non-replicated setups. What > is not clear is how much more data. This not only affects our network > bandwidth estimations, but also I/O load on the master server, slowing > writes (and some reads that cannot happen on the slave). Our database has about 2MB/s writes on the WAL array, we had about 160 IOPS in average when replications was switched off, and 165-170 IOPS in replication. This I think could be explained with statistical error, so we have not experienced any I/O load on our master server since replication was configured. -- Best regards Aleksej Trofimov