From mboxrd@z Thu Jan 1 00:00:00 1970 From: Allen Samuels Subject: RE: Adding compression/checksum support for bluestore. Date: Wed, 30 Mar 2016 22:35:18 +0000 Message-ID: References: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 8BIT Return-path: Received: from mail-bn1bon0096.outbound.protection.outlook.com ([157.56.111.96]:42228 "EHLO na01-bn1-obe.outbound.protection.outlook.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1751212AbcC3WfW convert rfc822-to-8bit (ORCPT ); Wed, 30 Mar 2016 18:35:22 -0400 In-Reply-To: Content-Language: en-US Sender: ceph-devel-owner@vger.kernel.org List-ID: To: Sage Weil , Vikas Sinha-SSI Cc: Igor Fedotov , ceph-devel > -----Original Message----- > From: Sage Weil [mailto:sage@newdream.net] > Sent: Wednesday, March 30, 2016 3:25 PM > To: Vikas Sinha-SSI > Cc: Allen Samuels ; Igor Fedotov > ; ceph-devel > Subject: RE: Adding compression/checksum support for bluestore. > > On Wed, 30 Mar 2016, Vikas Sinha-SSI wrote: > > > [snip] > > > > If I understand correctly, then there would still be a cost associated > > with writing dis-contiguously to disk. In cases such as this where the > > resources for compression are not easily available, I wonder if it is > > reasonable to simply not do compression for that Write. The cost of not > compressing would be a missed space optimization, but the cost of > compressing in any and all cases could be significant to latency. > > I also like this better than the split pextent complexity. I agree that the complexity is reduced -- which is good. Of course this is traded-off with a reduced storage capability, provided that this doesn't happen very often -- this might be a reasonable approach . I want to brood on this for a while. > > sage