From mboxrd@z Thu Jan 1 00:00:00 1970 From: Wols Lists Subject: Re: [RFC PATCH 0/7] evacuate struct page from the block layer Date: Fri, 20 Mar 2015 21:17:54 +0000 Message-ID: <550C8E82.2020309@youngman.org.uk> References: <20150316201640.33102.33761.stgit@dwillia2-desk3.amr.corp.intel.com> <20150318132650.3336261c58829f49a9af8675@linux-foundation.org> <20150319134313.GF4003@linux.intel.com> <550C490E.1080708@redhat.com> <20150320203136.GM4003@linux.intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20150320203136.GM4003@linux.intel.com> Sender: linux-kernel-owner@vger.kernel.org To: Matthew Wilcox , Rik van Riel Cc: Andrew Morton , Dan Williams , linux-kernel@vger.kernel.org, linux-arch@vger.kernel.org, axboe@kernel.dk, linux-nvdimm@ml01.01.org, Dave Hansen , linux-raid@vger.kernel.org, mgorman@suse.de, hch@infradead.org, linux-fsdevel@vger.kernel.org, "Michael S. Tsirkin" List-Id: linux-arch.vger.kernel.org On 20/03/15 20:31, Matthew Wilcox wrote: > Ah! I've looked at that a couple of times as well. I asked our database > performance team what impact freeing up the memmap would have on their > performance. They told me that doubling the amount of memory generally > resulted in approximately a 40% performance improvement. So freeing up > 1.5% additional memory would result in about 0.6% performance improvement, > which I thought was probably too small a return on investment to justify > turning memmap into a two-level data structure. Don't get me started on databases! This is very much a relational problem, other databases don't suffer like this. (imho relational theory is totally inappropriate for an engineering problem, like designing a database engine ...) Cheers, Wol From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from auth-4.ukservers.net ([217.10.138.158]:44361 "EHLO auth-4.ukservers.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751310AbbCTVR6 (ORCPT ); Fri, 20 Mar 2015 17:17:58 -0400 Message-ID: <550C8E82.2020309@youngman.org.uk> Date: Fri, 20 Mar 2015 21:17:54 +0000 From: Wols Lists MIME-Version: 1.0 Subject: Re: [RFC PATCH 0/7] evacuate struct page from the block layer References: <20150316201640.33102.33761.stgit@dwillia2-desk3.amr.corp.intel.com> <20150318132650.3336261c58829f49a9af8675@linux-foundation.org> <20150319134313.GF4003@linux.intel.com> <550C490E.1080708@redhat.com> <20150320203136.GM4003@linux.intel.com> In-Reply-To: <20150320203136.GM4003@linux.intel.com> Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: 7bit Sender: linux-arch-owner@vger.kernel.org List-ID: To: Matthew Wilcox , Rik van Riel Cc: Andrew Morton , Dan Williams , linux-kernel@vger.kernel.org, linux-arch@vger.kernel.org, axboe@kernel.dk, linux-nvdimm@ml01.01.org, Dave Hansen , linux-raid@vger.kernel.org, mgorman@suse.de, hch@infradead.org, linux-fsdevel@vger.kernel.org, "Michael S. Tsirkin" Message-ID: <20150320211754.9CXlxS9ustyjK9B4PIkTrHLPp4XfpBy4-ESpO0tP41I@z> On 20/03/15 20:31, Matthew Wilcox wrote: > Ah! I've looked at that a couple of times as well. I asked our database > performance team what impact freeing up the memmap would have on their > performance. They told me that doubling the amount of memory generally > resulted in approximately a 40% performance improvement. So freeing up > 1.5% additional memory would result in about 0.6% performance improvement, > which I thought was probably too small a return on investment to justify > turning memmap into a two-level data structure. Don't get me started on databases! This is very much a relational problem, other databases don't suffer like this. (imho relational theory is totally inappropriate for an engineering problem, like designing a database engine ...) Cheers, Wol