From mboxrd@z Thu Jan 1 00:00:00 1970 From: xiao xianxia Subject: Re: location-aware file placement in Ceph Date: Mon, 14 Apr 2014 09:27:24 +0800 Message-ID: <534B397C.30103@ubuntukylin.com> References: <53457C07.3060507@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Received: from m199-177.yeah.net ([123.58.177.199]:43775 "EHLO m199-177.yeah.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750938AbaDNBg1 (ORCPT ); Sun, 13 Apr 2014 21:36:27 -0400 In-Reply-To: <53457C07.3060507@gmail.com> Sender: ceph-devel-owner@vger.kernel.org List-ID: To: Zhongliang Zhao Cc: ceph-devel@vger.kernel.org, liwang Hi Zhongliang, you can find a lot of command on ceph official document! =E4=BA=8E 2014/4/10 0:57, Zhongliang Zhao =E5=86=99=E9=81=93: > Hi XIanxia, > > Thanks a lot for the reply. > > Since I am new to ceph, some parts of your reply is unclear to me. Is= =20 > it possible for you to make it more clear? Thanks a lot for your time= =2E > > On 02/28/2014 02:41 AM, Xianxia Xiao wrote: >> >> May be able to meet your needs through rbd. >> >> >> You can create multiple rules,and then specify a different pool to=20 >> use a different ruleset. > Where can I modify or creat the rule? On the monitor node? and how to= =20 > configure a pool to use a ruleset? you can use crushtool export crush map,then modify the rule in crush ma= p=20 ,you can create new rules. http://ceph.com/docs/master/rados/operations/crush-map/?highlight=3Dcru= sh%20map you can use #ceph osd pool set {pool name} {key} {value} the key is crush_ruleset,value is the rule num. >> Then create multiple block images, and Specify image to different po= ols. > How this can be done? you can see that in Block Device Command in ceph document. http://ceph.com/docs/master/rbd/rados-rbd-cmds/ >> create file system on the block device and mount file system on=20 >> different positions,so that the different location of the file can=20 >> use different ruleset. > How do I know to where I should mount the file system? You decide,mount command is the same to linux mount command. > > Best regards, > Zhongliang Zhao >> >> >> Best regards, xian xia >> >> >> >> >> >>> -----=E5=8E=9F=E5=A7=8B=E9=82=AE=E4=BB=B6----- >>> =E5=8F=91=E4=BB=B6=E4=BA=BA: Zhongliang Zhao >>> =E5=8F=91=E9=80=81=E6=97=B6=E9=97=B4: 2014=E5=B9=B42=E6=9C=8827=E6=97= =A5 =E6=98=9F=E6=9C=9F=E5=9B=9B >>> =E6=94=B6=E4=BB=B6=E4=BA=BA: ceph-devel@vger.kernel.org >>> =E6=8A=84=E9=80=81: >>> =E4=B8=BB=E9=A2=98: location-aware file placement in Ceph >>> >>> Dear all, >>> >>> As new to ceph, may I ask in the crush algorithm, is there a way to >>> configure where the file to be stored, not randomly. >>> My requirement is to place the files that within the same >>> location(within a geographic area), into the same OSD, instead of >>> randomly placing them seperately. >>> >>> I guess crush may support this. But to do this, I just have to=20 >>> configure >>> some parameters in some ceph commands? or I have to modify the ceph >>> crush algorithm and re-compile ceph? >>> >>> Many thanks for any helpful pointers. >>> >>> Cheers, Giovanni >>> --=20 >>> To unsubscribe from this list: send the line "unsubscribe=20 >>> ceph-devel" in >>> the body of a message to majordomo@vger.kernel.org >>> More majordomo info at http://vger.kernel.org/majordomo-info.html >> > -- 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