Re: [Xen-users] lots of cycles in i/o wait state

[Miles Fidelman <mfidelman@xxxxxxxxxxxxxxxx>]

Pasi Kärkkäinen wrote:
> On Sat, Jun 05, 2010 at 06:59:51PM -0400, Miles Fidelman wrote:
>    
> > Hi Folks,
> >
> > I've been doing some experimenting to see how far I can push some old
> > hardware into a virtualized environment - partially to see how much use
> > I can get out of the hardware, and partially to learn more about the
> > behavior of, and interactions between, software RAID, LVM, DRBD, and Xen.
> >
> >      
> Is your disk/partition aligment properly set up? Doing it wrong could
> cause bad performance. It's easy to mess it up with VMs.
>    
can you say a little more about what you mean by "properly set up" vs. 
not properly set up?
> > As I've started experimenting with adding additional domUs, in various
> > configurations, I've found that my mail server can get into a state
> > where it's spending almost all of its cycles in an i/o wait state (95%
> > and higher as reported by top).  This is particularly noticeable when I
> > run a backup job (essentially a large tar job that reads from the root
> > volume and writes to the backup volume).  The domU grinds to halt.
> >      
> Is that iowait measure in the guest, or in dom0?
>    
iowait ONLY suffers in the guest

when I run stress tests, iowait (in the guest) jumps considerably when:
- running a benchmark (bonnie++) in dom0, on either host (to be 
expected, given that dom0 gets priority)
- running bonnie++ in the guest with iowait problems

running bonnie++ in another guest does not impact the iowaits
> Again run "iostat 1" in both the domU and dom0, and compare the results.
> Also run "xm top" in dom0 to monitor the overall CPU usage.
>    
very little CPU load

iostat (and vmstat) are what really helped me track things down; and 
after doing a lot of googling on "performance tuning" and "iowait" I 
came across the suggestion to add "noatime" to my mount options --- 
brought my iowait times way down, and sped up performance
you learn something new every day :-)

Thanks again, to all,

Miles Fidelman




--
In theory, there is no difference between theory and practice.
In<fnord>  practice, there is.   .... Yogi Berra



_______________________________________________
Xen-users mailing list
Xen-users@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-users