Disregard I just saw it was a dd option.

Simone

> -----Original Message-----
> From: open-iscsi@googlegroups.com [mailto:open-is...@googlegroups.com] On
> Behalf Of Simone Morellato
> Sent: Monday, April 13, 2009 3:04 PM
> To: open-iscsi@googlegroups.com
> Subject: RE: Open iSCSI Performance on IBM
> 
> 
> Hi Mike,
> 
> Is bs=128K a linux, iscsi or IBM parameter?
> 
> Thanks,
> Simone
> 
> 
> > -----Original Message-----
> > From: open-iscsi@googlegroups.com [mailto:open-is...@googlegroups.com]
> On
> > Behalf Of Mike Christie
> > Sent: Thursday, April 09, 2009 10:55 AM
> > To: open-iscsi@googlegroups.com
> > Subject: Re: Open iSCSI Performance on IBM
> >
> >
> > Gonçalo Borges wrote:
> > > Hi All...
> > >
> > > Sorry, the following could be a little bit off topic...
> > >
> > > Does any one has an idea of what is the expected performance for a IBM
> > > DS 3300 system connected via open iSCSI? Using a RAID 1 with 2 disks,
> > > I got the following numbers:
> > >
> >
> > Is /apoio04/b1 a scsi/iscsi disk or is it LVM/DM/RAID on top of a
> > iscsi/scsi disk?
> >
> > Could you set the IO scheduler to noop
> > echo noop > /sys/block/sdX/queue/scheduler
> >
> > and see if that makes a difference.
> >
> > Also try
> >
> > bs=128k
> >
> > And then also run
> > iscsiadm -m session -P 3
> >
> >
> >
> > > Sequential Write:
> > >
> > >  [r...@core12 ~]# dd if=/dev/zero of=/apoio04/b1 bs=64k count=125000
> > > 125000+0 records in
> > > 125000+0 records out
> > > 8192000000 bytes (8.2 GB) copied, 454.522 seconds, 18.0 MB/s
> > >
> > > Sequential Read:
> > > [r...@core12 ~]# dd if=/apoio04/b1 of=/dev/null bs=64k count=125000
> > > 125000+0 records in
> > > 125000+0 records out
> > > 8192000000 bytes (8.2 GB) copied, 94.9401 seconds, 86.3 MB/s
> > >
> > > I restricted the RAM to be only 1GB, therefore there are no cache
> > > effects in these numbers. Because the read stats are good, we exclude
> > > network bottlenecks. Nevertheless, we were expecting more or less the
> > > performance of a single disk (~50MB/s) for the write tests and we are
> > > getting less than half. I do not know if this is really the physical
> > > limit of the system or if there is a problem somewhere...
> > >
> > > I could not find any IBM official numbers, therefore, I though that
> > > someone over here could give me any hint about the numbers they are
> > > getting...
> > >
> > > Thanks in Advance
> > > Cheers
> > > Goncalo Borges
> > > >
> >
> >
> >
> 
> 

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"open-iscsi" group.
To post to this group, send email to open-iscsi@googlegroups.com
To unsubscribe from this group, send email to 
open-iscsi+unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/open-iscsi
-~----------~----~----~----~------~----~------~--~---

Reply via email to