[Top] [All Lists]

Re: XFS - Performance Optimazation.

To: "C.G.Senthilkumar." <senthilkumar.cheeta@xxxxxxxxx>
Subject: Re: XFS - Performance Optimazation.
From: "Martin K. Petersen" <mkp@xxxxxxxxxxxxx>
Date: 25 Apr 2001 11:28:11 -0400
Cc: linux-xfs@xxxxxxxxxxx
In-reply-to: <200104251515.f3PFFnb24856@jen.americas.sgi.com>
Organization: Linuxcare, Inc.
References: <200104251515.f3PFFnb24856@jen.americas.sgi.com>
Sender: owner-linux-xfs@xxxxxxxxxxx
User-agent: Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.2 (Urania)
>>>>> "Steve" == Steve Lord <lord@xxxxxxx> writes:

Steve> OK, now to the crux of the matter, these mkfs options are not
Steve> going to help you achieve higher bandwidth, your hardware is
Steve> going to play the biggest part in this, so it is all really a
Steve> matter of budget. We have had a single file doing streaming I/O
Steve> at close to 100 Mbytes/sec on linux, however, that involved a
Steve> fiber channel connected JBOD running lvm on 8 10000 rpm scsi
Steve> drives. 

MD RAID0, actually.  I get slightly lower performance with LVM.

Steve> You should probably build an lvm volume striped across several
Steve> drives, the best stripe width is probably going to depend on
Steve> how the application does I/O, but it depends on how much
Steve> bandwidth you need to squeeze out.

Modern drives seem to like 64-128KB I/Os.  The test Steve mentioned
above was done with a 64KB stripe unit and consequently 512 KB stripe

Martin K. Petersen, Principal Linux Consultant, Linuxcare, Inc.
mkp@xxxxxxxxxxxxx, http://www.linuxcare.com/
SGI XFS for Linux Developer, http://oss.sgi.com/projects/xfs/

<Prev in Thread] Current Thread [Next in Thread>