From mboxrd@z Thu Jan 1 00:00:00 1970 From: Mark Nelson Subject: Re: Ceph Production Environment Setup and Configurations? Date: Tue, 29 Jan 2013 07:40:23 -0600 Message-ID: <5107D147.5020708@inktank.com> References: <6F3FA899187F0043BA1827A69DA2F7CC5E93B4@SHSMSX102.ccr.corp.intel.com> <5107B3D8.8060504@inktank.com> Mime-Version: 1.0 Content-Type: text/plain; charset=GB2312 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Received: from mail-ia0-f171.google.com ([209.85.210.171]:36964 "EHLO mail-ia0-f171.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751801Ab3A2Nk0 (ORCPT ); Tue, 29 Jan 2013 08:40:26 -0500 Received: by mail-ia0-f171.google.com with SMTP id z13so557454iaz.16 for ; Tue, 29 Jan 2013 05:40:25 -0800 (PST) In-Reply-To: <5107B3D8.8060504@inktank.com> Sender: ceph-devel-owner@vger.kernel.org List-ID: To: Joao Eduardo Luis Cc: "Chen, Xiaoxi" , Gandalf Corvotempesta , femi anjorin , "ceph-devel@vger.kernel.org" , Ross Turk On 01/29/2013 05:34 AM, Joao Eduardo Luis wrote: > On 01/29/2013 11:25 AM, Chen, Xiaoxi wrote: >> [The following views only behalf of myself, not relate with Intel..] >> Looking forward for the performance data on Atom. >> Atom perform badly in Swift, but since Ceph is slightly efficient th= an Swift, it must be better. >> I have some concern about weather Atom can support such high through= put( you have 16 disks, assuming 50MB/s per disk, you would like to hav= e 50*16 *2(include journal write)=3D1600MB/s ,together with correspondi= ng network throughput, say 10GbE. The total IO throughput seems too hig= h for Atom. Baidu (www.baidu.com) use Arm based storage node(not using = ceph,but their own DFS), a node with a quad-core Arm together with 4 di= sks, and a 2U box can put up to 6 nodes, the 6 nodes share a 10Gb NIC. = Although Atom is different with Arm, but you can take Baidu as a refere= nce. >> >> >> -----Original Message----- >> From: ceph-devel-owner@vger.kernel.org [mailto:ceph-devel-owner@vger= =2Ekernel.org] On Behalf Of Gandalf Corvotempesta >> Sent: 2013=C4=EA1=D4=C229=C8=D5 17:25 >> To: femi anjorin >> Cc: ceph-devel@vger.kernel.org; Ross Turk >> Subject: Re: Ceph Production Environment Setup and Configurations? >> >> 2013/1/29 femi anjorin : >>> CPU - Intel(R) Atom(TM) CPU D525 @ 1.80GHz >> >> Atom? For which kind of role ? >> -- >> To unsubscribe from this list: send the line "unsubscribe ceph-devel= " in the body of a message to majordomo@vger.kernel.org More majordomo = info at http://vger.kernel.org/majordomo-info.html >> N=8B=A7=B2=E6=ECr=B8=9By=FA=E8=9A=D8b=B2X=AC=B6=C7=A7v=D8^=81=7F=7F)= =DE=BA{.n=81=7F=7F+=89=B7=9Cz=98]z=F7=A5=8A{ay=81=7F=7F=1D=CA=87=DA=99=81= =7F=7F,j=07=AD=A2f=A3=A2=B7h=9A=8B=E0z=81=7F=7F=1E=AEw=A5=A2=81=7F=7F=0C= =A2=B7=A6j:+v=89=A8=8Aw=E8j=D8m=B6=9F=81=7F=7F=81=7F=7F=07=AB=91=EA=E7z= Z+=83=F9=9A=8E=8A=DD=A2j"=9D=FA!tml=3D >> >=20 >=20 > FWIW, Wido had been testing Ceph on Atoms a while back. I don't know > what conclusions he reached, or even if he has already reached any > conclusions at all. I, for one, would be really interested in knowin= g > how well (or how poorly) Atoms deal with Ceph. >=20 > Maybe Wido can share some thoughts on this one? :-) >=20 > -Joao > -- > To unsubscribe from this list: send the line "unsubscribe ceph-devel"= in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html >=20 Just FYI, I don't expect you'll be able to get anywhere close to 10GbE performance on the Atom. You'll almost certainly be CPU bound long before that. I actually would be curious how much throughput you could push to those disks just doing straight fio tests and how much CPU overhead you'd see. I imagine you could max out the CPU without even having Ceph involved. I'm guessing you probably are going to be limited to about 0.75-1 OSD per atom core, and that's assuming that your network and SAS controller= s aren't CPU hogs. Mark -- 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