RE: [Xen-ia64-devel] [PATCH] [Resend]Enable hash vtlb

[Alex Williamson <alex.williamson@xxxxxx>]

On Mon, 2006-04-10 at 23:01 +0800, Xu, Anthony wrote:

> If we configure domU with memory 256MB, domU will complain "at least 256M 
> is needed." 
> Yes there should a best ratio of memory size of domU and size of VHPT.

My tests are:

dom0: boot w/ dom0_mem=768M, kill off all daemons, build
domU: boot w/ default dom0 mem (512MB), kill all daemons in dom0,
specify 768M memory from domU, boot domU, kill all domU daemons, build

256MB certainly isn't enough memory to have a worthwhile kernel build
benchmark.

> >   I don't understand this result.  I was surprised to see domU perform
> >better than dom0 in my testing, but I can't see how domU could perform
> >better than bare metal.  Perhaps 512MB is insufficient for kernel
> >builds.  You may be disproportionately benefiting from dom0's buffer
> >cache.
> >
> I think there maybe two reasons.
> 1. As you said, domU benefits from dom0's buffer cache. There are somewhat 
> parallel executions. DomU is response of compilation, Dom0 is response of 
> read/write of disk.
> 2. The services running on Dom0 or DomU are less than that on native machine.

   Services can also be stopped on the native machine.  I did this in my
test case.  I think it's very possible that 512MB is not a sufficient
amount of memory for a valid test.  768MB may not be enough either.  To
properly benchmark this change we need to have the entire working set of
the test fit in memory (preferably we'd do the builds out of a tmpfs
mount to avoid I/O entirely).  If we have extra activity, like swapping
or text getting pushed out of buffer cache and reloaded, anything we can
read into the results is suspect.  Thanks,

        Alex

-- 
Alex Williamson                             HP Linux & Open Source Lab


_______________________________________________
Xen-ia64-devel mailing list
Xen-ia64-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-ia64-devel