From mboxrd@z Thu Jan 1 00:00:00 1970 From: Zhi Yong Wu Subject: Re: [Qemu-devel] [PATCH v4 0/3] The intro for QEMU disk I/O limits Date: Fri, 5 Aug 2011 10:20:48 +0800 Message-ID: References: <1312179955-23536-1-git-send-email-wuzhy@linux.vnet.ibm.com> <20110801200622.GL7358@us.ibm.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: Zhi Yong Wu , qemu-devel@nongnu.org, kwolf@redhat.com, aliguori@us.ibm.com, stefanha@linux.vnet.ibm.com, kvm@vger.kernel.org, mtosatti@redhat.com, luowenj@cn.ibm.com, raharper@us.ibm.com To: Ryan Harper Return-path: Received: from mail-gx0-f174.google.com ([209.85.161.174]:46431 "EHLO mail-gx0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754411Ab1HECUu convert rfc822-to-8bit (ORCPT ); Thu, 4 Aug 2011 22:20:50 -0400 Received: by gxk21 with SMTP id 21so1444663gxk.19 for ; Thu, 04 Aug 2011 19:20:50 -0700 (PDT) In-Reply-To: <20110801200622.GL7358@us.ibm.com> Sender: kvm-owner@vger.kernel.org List-ID: On Tue, Aug 2, 2011 at 4:06 AM, Ryan Harper wrote: > * Zhi Yong Wu [2011-08-01 01:30]: >> The main goal of the patch is to effectively cap the disk I/O speed = or counts of one single VM.It is only one draft, so it unavoidably has = some drawbacks, if you catch them, please let me know. >> >> The patch will mainly introduce one block I/O throttling algorithm, = one timer and one block queue for each I/O limits enabled drive. >> >> When a block request is coming in, the throttling algorithm will che= ck if its I/O rate or counts exceed the limits; if yes, then it will en= queue to the block queue; The timer will handle the I/O requests in it. >> >> Some available features follow as below: >> (1) global bps limit. >> =A0 =A0 -drive bps=3Dxxx =A0 =A0 =A0 =A0 =A0 =A0in bytes/s >> (2) only read bps limit >> =A0 =A0 -drive bps_rd=3Dxxx =A0 =A0 =A0 =A0 in bytes/s >> (3) only write bps limit >> =A0 =A0 -drive bps_wr=3Dxxx =A0 =A0 =A0 =A0 in bytes/s >> (4) global iops limit >> =A0 =A0 -drive iops=3Dxxx =A0 =A0 =A0 =A0 =A0 in ios/s >> (5) only read iops limit >> =A0 =A0 -drive iops_rd=3Dxxx =A0 =A0 =A0 =A0in ios/s >> (6) only write iops limit >> =A0 =A0 -drive iops_wr=3Dxxx =A0 =A0 =A0 =A0in ios/s >> (7) the combination of some limits. >> =A0 =A0 -drive bps=3Dxxx,iops=3Dxxx >> >> Known Limitations: >> (1) #1 can not coexist with #2, #3 >> (2) #4 can not coexist with #5, #6 >> (3) When bps/iops limits are specified to a small value such as 511 = bytes/s, this VM will hang up. We are considering how to handle this se= nario. >> >> >> Zhi Yong Wu (3): >> =A0 v4: fix memory leaking based on ryan's feedback. > > It looks like the leak has been fixed, but I think we need to rework = how > the blk-queue is using the AIOPool. =A0I'll reply to that patch. OK. Look forward to seeing it. > > > > -- > Ryan Harper > Software Engineer; Linux Technology Center > IBM Corp., Austin, Tx > ryanh@us.ibm.com > --=20 Regards, Zhi Yong Wu From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([140.186.70.92]:49085) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1QpA1s-00060w-BV for qemu-devel@nongnu.org; Thu, 04 Aug 2011 22:20:53 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1QpA1r-0007z1-4Z for qemu-devel@nongnu.org; Thu, 04 Aug 2011 22:20:52 -0400 Received: from mail-gx0-f173.google.com ([209.85.161.173]:45155) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1QpA1q-0007yx-V1 for qemu-devel@nongnu.org; Thu, 04 Aug 2011 22:20:51 -0400 Received: by gxk26 with SMTP id 26so1690561gxk.4 for ; Thu, 04 Aug 2011 19:20:50 -0700 (PDT) MIME-Version: 1.0 In-Reply-To: <20110801200622.GL7358@us.ibm.com> References: <1312179955-23536-1-git-send-email-wuzhy@linux.vnet.ibm.com> <20110801200622.GL7358@us.ibm.com> Date: Fri, 5 Aug 2011 10:20:48 +0800 Message-ID: From: Zhi Yong Wu Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Subject: Re: [Qemu-devel] [PATCH v4 0/3] The intro for QEMU disk I/O limits List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Ryan Harper Cc: kwolf@redhat.com, aliguori@us.ibm.com, stefanha@linux.vnet.ibm.com, kvm@vger.kernel.org, mtosatti@redhat.com, qemu-devel@nongnu.org, Zhi Yong Wu , luowenj@cn.ibm.com, raharper@us.ibm.com On Tue, Aug 2, 2011 at 4:06 AM, Ryan Harper wrote: > * Zhi Yong Wu [2011-08-01 01:30]: >> The main goal of the patch is to effectively cap the disk I/O speed or c= ounts of one single VM.It is only one draft, so it unavoidably has some dra= wbacks, if you catch them, please let me know. >> >> The patch will mainly introduce one block I/O throttling algorithm, one = timer and one block queue for each I/O limits enabled drive. >> >> When a block request is coming in, the throttling algorithm will check i= f its I/O rate or counts exceed the limits; if yes, then it will enqueue to= the block queue; The timer will handle the I/O requests in it. >> >> Some available features follow as below: >> (1) global bps limit. >> =A0 =A0 -drive bps=3Dxxx =A0 =A0 =A0 =A0 =A0 =A0in bytes/s >> (2) only read bps limit >> =A0 =A0 -drive bps_rd=3Dxxx =A0 =A0 =A0 =A0 in bytes/s >> (3) only write bps limit >> =A0 =A0 -drive bps_wr=3Dxxx =A0 =A0 =A0 =A0 in bytes/s >> (4) global iops limit >> =A0 =A0 -drive iops=3Dxxx =A0 =A0 =A0 =A0 =A0 in ios/s >> (5) only read iops limit >> =A0 =A0 -drive iops_rd=3Dxxx =A0 =A0 =A0 =A0in ios/s >> (6) only write iops limit >> =A0 =A0 -drive iops_wr=3Dxxx =A0 =A0 =A0 =A0in ios/s >> (7) the combination of some limits. >> =A0 =A0 -drive bps=3Dxxx,iops=3Dxxx >> >> Known Limitations: >> (1) #1 can not coexist with #2, #3 >> (2) #4 can not coexist with #5, #6 >> (3) When bps/iops limits are specified to a small value such as 511 byte= s/s, this VM will hang up. We are considering how to handle this senario. >> >> >> Zhi Yong Wu (3): >> =A0 v4: fix memory leaking based on ryan's feedback. > > It looks like the leak has been fixed, but I think we need to rework how > the blk-queue is using the AIOPool. =A0I'll reply to that patch. OK. Look forward to seeing it. > > > > -- > Ryan Harper > Software Engineer; Linux Technology Center > IBM Corp., Austin, Tx > ryanh@us.ibm.com > --=20 Regards, Zhi Yong Wu