Received: from localhost (unknown [200.46.204.183]) by postgresql.org (Postfix) with ESMTP id D7CE264FCC6 for ; Mon, 15 Sep 2008 11:26:08 -0300 (ADT) Received: from postgresql.org ([200.46.204.86]) by localhost (mx1.hub.org [200.46.204.183]) (amavisd-maia, port 10024) with ESMTP id 75675-01 for ; Mon, 15 Sep 2008 11:26:05 -0300 (ADT) X-Greylist: from auto-whitelisted by SQLgrey-1.7.6 Received: from relay-1.mail.nethere.net (relay-1.mail.nethere.net [66.63.128.161]) by postgresql.org (Postfix) with ESMTP id 67BF864FCBB for ; Mon, 15 Sep 2008 11:26:05 -0300 (ADT) Received: from scan-2.mail.nethere.net (scan-2.mail.nethere.net [66.63.128.133]) by relay-1.mail.nethere.net (Postfix) with ESMTP id 1EB2C29180; Mon, 15 Sep 2008 07:26:03 -0700 (PDT) Received: from mta-1.mail.nethere.net by scan-2.mail.nethere.net with LMTP id VEhvnf71awrW; Mon, 15 Sep 2008 07:26:03 -0700 (PDT) Received: from lisa-james-computer.local (wsip-68-224-174-233.sd.sd.cox.net [68.224.174.233]) by mta-1.mail.nethere.net (Postfix) with ESMTP id C7008F1818; Mon, 15 Sep 2008 07:26:02 -0700 (PDT) Message-ID: <48CE7076.1040309@emolecules.com> Date: Mon, 15 Sep 2008 07:25:58 -0700 From: Craig James User-Agent: Thunderbird 2.0.0.16 (Macintosh/20080707) MIME-Version: 1.0 To: Florian Weimer CC: PostgreSQL Performance , Matthew Wakeling Subject: Re: select on 22 GB table causes "An I/O error occured while sending to the backend." exception References: <82bpze3ou8.fsf@mid.bfk.de> <48B6EA7E.2090707@emolecules.com> <82d4j5ls9z.fsf@mid.bfk.de> <48CE6D0D.80509@emolecules.com> <82ej3ljy0l.fsf@mid.bfk.de> In-Reply-To: <82ej3ljy0l.fsf@mid.bfk.de> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Virus-Scanned: Maia Mailguard 1.0.1 X-Spam-Status: No, hits=0 tagged_above=0 required=5 tests=none X-Spam-Level: X-Archive-Number: 200809/173 X-Sequence-Number: 31486 Florian Weimer wrote: > * Craig James: > >>> There are some run-time environments which allocate huge chunks of >>> memory on startup, without marking them as not yet in use. SBCL is in >>> this category, and also the Hotspot VM (at least some extent). >> I stand by my assertion: It never makes sense. Do these >> applications allocate a terrabyte of memory? I doubt it. > > SBCL sizes its allocated memory region based on the total amount of > RAM and swap space. In this case, buying larger disks does not > help. 8-P SBCL, as Steel Bank Common Lisp? Why would you run that on a server machine alongside Postgres? If I had to use SBLC and Postgres, I'd put SBLC on a separate machine all its own, so that it couldn't corrupt Postgres or other servers that had to be reliable. Are you saying that if I bought a terrabyte of swap disk, SBLC would allocate a terrabyte of space? Craig