From mboxrd@z Thu Jan 1 00:00:00 1970 From: Vladislav Bolkhovitin Subject: Re: Linux I/O subsystem performance (was: linuxcon 2010...) Date: Tue, 24 Aug 2010 18:43:29 +0400 Message-ID: <4C73DA91.9040101@vlnb.net> References: <8A96806D-6CD7-44AD-8A9D-143C098C95A4@uni-paderborn.de> <1282256949.30453.278.camel@haakon2.linux-iscsi.org> <4C701E08.2020005@vlnb.net> <1282423398.3015.39.camel@mulgrave.site> <1282508953.3042.102.camel@mulgrave.site> <4C727BEB.9020100@scalableinformatics.com> <20100824072557.GK2804@reaktio.net> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Received: from moutng.kundenserver.de ([212.227.17.9]:62167 "EHLO moutng.kundenserver.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752071Ab0HXOnY (ORCPT ); Tue, 24 Aug 2010 10:43:24 -0400 In-Reply-To: <20100824072557.GK2804@reaktio.net> Sender: linux-scsi-owner@vger.kernel.org List-Id: linux-scsi@vger.kernel.org To: =?ISO-8859-1?Q?Pasi_K=E4rkk=E4inen?= Cc: Chetan Loke , Bart Van Assche , linux-scsi@vger.kernel.org, linux-kernel@vger.kernel.org, James Bottomley , scst-devel Pasi K=E4rkk=E4inen, on 08/24/2010 11:25 AM wrote: > On Mon, Aug 23, 2010 at 02:03:26PM -0400, Chetan Loke wrote: >> I actually received 3+ off-post emails asking whether I was talking >> about initiator or target in the 100K IOPS case below and what did I >> mean by the ACKs. >> I was referring to the 'Initiator' side. >> ACKs =3D=3D When scsi-ML down-calls the LLD via the queue-command, p= rocess >> the sgl's(if you like) and then trigger the scsi_done up-call path. >> > > Uhm, Intel and Microsoft demonstrated over 1 million IOPS > using software iSCSI and a single 10 Gbit Ethernet NIC (Intel 82599). > > How come there is such a huge difference? What are we lacking in Linu= x? I also have an impression that Linux I/O subsystem has some performance= =20 problems. For instance, in one recent SCST performance test only 8 Linu= x=20 initiators with fio as a load generator were able to saturate a single=20 SCST target with dual IB cards (SRP) on 4K AIO direct accesses over an=20 SSD backend. This rawly means that any initiator took several times (8?= )=20 more processing time than the target. Hardware used for that target and= =20 initiators was the same. I can't see on this load why the initiators=20 would need to do something more than the target. Well, I know we in SCS= T=20 did an excellent work to maximize performance, but such a difference=20 looks too much ;) Also it looks very suspicious why nobody even tried to match that=20 Microsoft/Intel record, even Intel itself who closely works with Linux=20 community in the storage area and could do it using the same hardware. Vlad -- To unsubscribe from this list: send the line "unsubscribe linux-scsi" i= n the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html