From mboxrd@z Thu Jan 1 00:00:00 1970 From: Alexandre DERUMIER Subject: Re: Severe performance degradation with jewel rbd image Date: Wed, 1 Jun 2016 08:58:30 +0200 (CEST) Message-ID: <1201586295.760757370.1464764310119.JavaMail.zimbra@oxygem.tv> References: <417572431.751254216.1464420555984.JavaMail.zimbra@oxygem.tv> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Received: from mailpro.odiso.net ([89.248.211.110]:55937 "EHLO mailpro.odiso.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750697AbcFAG6d convert rfc822-to-8bit (ORCPT ); Wed, 1 Jun 2016 02:58:33 -0400 In-Reply-To: Sender: ceph-devel-owner@vger.kernel.org List-ID: To: dillaman@redhat.com Cc: Somnath Roy , Samuel Just , ceph-devel >>Once librbd/librados can eliminate all the=20 >>threading/queuing slowdowns, a similar approach could be extended to=20 >>RBD images without the need to open multiple copies of the same image= =2E=20 Thanks Jason, that's a really good news ! ----- Mail original ----- De: "Jason Dillaman" =C3=80: "aderumier" Cc: "Somnath Roy" , "Samuel Just" , "ceph-devel" Envoy=C3=A9: Mardi 31 Mai 2016 21:57:29 Objet: Re: Severe performance degradation with jewel rbd image That QEMU patch looks like it only supports raw devices and=20 pre-allocated files. Once librbd/librados can eliminate all the=20 threading/queuing slowdowns, a similar approach could be extended to=20 RBD images without the need to open multiple copies of the same image.=20 On Sat, May 28, 2016 at 3:29 AM, Alexandre DERUMIER wrote:=20 > Hi,=20 >=20 > qemu should have soon multithreaded disk access support (multi queues= + multi iothreads)=20 >=20 > https://lists.gnu.org/archive/html/qemu-devel/2016-05/msg05023.html=20 >=20 > Do you think it'll work with exclusive-lock ?=20 >=20 > ----- Mail original -----=20 > De: "Jason Dillaman" =20 > =C3=80: "Somnath Roy" =20 > Cc: "Samuel Just" , "ceph-devel" =20 > Envoy=C3=A9: Jeudi 26 Mai 2016 20:52:26=20 > Objet: Re: Severe performance degradation with jewel rbd image=20 >=20 > You can dynamically enable/disable exclusive-lock, object-map,=20 > fast-diff, and journaling after the image is created via "rbd feature= =20 > disable ". The deep-flatten feature can be=20 > disabled-only via the same CLI.=20 > --=20 > To unsubscribe from this list: send the line "unsubscribe ceph-devel"= in=20 > the body of a message to majordomo@vger.kernel.org=20 > More majordomo info at http://vger.kernel.org/majordomo-info.html=20 > --=20 > To unsubscribe from this list: send the line "unsubscribe ceph-devel"= in=20 > the body of a message to majordomo@vger.kernel.org=20 > More majordomo info at http://vger.kernel.org/majordomo-info.html=20 --=20 Jason=20 -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" i= n the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html