[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: [Xen-devel] Does Xen hypervisor overwrite O_DIRECT setting of Linux2.6 kernel?

  • To: "Liang Yang" <multisyncfe991@xxxxxxxxxxx>, <xen-devel@xxxxxxxxxxxxxxxxxxx>
  • From: "Ian Pratt" <m+Ian.Pratt@xxxxxxxxxxxx>
  • Date: Sat, 28 Oct 2006 23:35:52 +0100
  • Delivery-date: Sat, 28 Oct 2006 15:39:59 -0700
  • List-id: Xen developer discussion <xen-devel.lists.xensource.com>
  • Thread-index: Acb63rX/jf1l52kcQUOv2nEeFVaCogAAcIAQ
  • Thread-topic: [Xen-devel] Does Xen hypervisor overwrite O_DIRECT setting of Linux2.6 kernel?

> Yes, all the testing configurations are the exactly same (both uses
> kernel and I used Xen 3.0.3. I did several indepedent runs
> my testing and the results are consistent, e.g. the performance of
> packet
> sequential write under Xen Domain0 outperform Linux native by 10~20%.
> Some people told me Xen hypervisor adds addtional layer beyond Linux
> stack. So O_DIRECT under Xen hypervisor is still valid, but I/O
> for small packets are enabled by Xen hypervisor. However, Linux native
> kernel would not do any I/O coalescing when using O_DIRECT.

For IO from dom0 there is no additional layer. There will be no
difference in the coalescing behaviour.

Are you sure you're writing to the same part of the disk in both cases?
You know about disk zoning, right?


> I hope Xen gurus can give me more explanations about this.
> Thanks,
> Liang
> ----- Original Message -----
> From: "Ian Pratt" <m+Ian.Pratt@xxxxxxxxxxxx>
> To: "Liang Yang" <multisyncfe991@xxxxxxxxxxx>;
> <xen-devel@xxxxxxxxxxxxxxxxxxx>
> Cc: <ian.pratt@xxxxxxxxxxxx>
> Sent: Saturday, October 28, 2006 2:34 PM
> Subject: RE: [Xen-devel] Does Xen hypervisor overwrite O_DIRECT
setting of
> Linux2.6 kernel?
> > The performance data is collected on 8 SAS drives (used as physical
> drives)
> > and IOMeter is used as the benchmark tool. The latest IOMeter
> used
> > O_DIRECT. We know, Linux 2.6 kernel starts supporting O_DIRECT which
> makes
> > all I/O requests work around buffer cache. The good thing for
> is
> > it
> > reduces the CPU utilization and cache pollution. The bad thing is
> > not only forces all I/O requests become synchronous and no I/O
> coalescing
> > will happen. Thus sequential write of small packets will be impacted
> most.
> > For Xen, however, I believe Xen hypervisor overwrites this O_DIRECT
> setting
> > and maybe it favors better performance over CPU and FSB utilization.
> Thus
> > Xen domain0 can have better write performance than Linux native.
> Are you sure you're comparing identical native and dom0 kernel
> Same drivers and settings?
> Xen does not disable O_DIRECT.
> Ian

Xen-devel mailing list



Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.