Opinions on packaging ATLAS (for the x86 architecture)

[Peter Robinson]

On Wed, Oct 7, 2009 at 4:29 PM, Bill Nottingham <notting at redhat.com> wrote:
> Richard W.M. Jones (rjones at redhat.com) said:
>> > This is going to be pretty important for scientific workloads where
>> > atlas is going to be used. I've eavesdropped on several conversations
>> > where people were talking about being able to run off-the-shelf
>> > science code virtual appliance in order to reduce the environment
>> > configuration workload for an individual researcher.
>>
>> Yup.  The really fun starts when you do live migration.  The processor
>> literally changes underneath the running programs.  If you thought you
>> had SSE3 one minute, then the next you don't, or vice versa.
>>
>> No one has to my knowledge come up with a good way to deal with this.
>> But it probably involves signalling the kernel and processes so that
>> they can redo processor detection.  You can see why that is not going
>> to be pleasant.
>
> Surely the way to do this is to know what your workload is doing,
> and not do live migration to random hardware?

In fact the proper way to do this it to have the same hardware in the
group of servers that VMs might be live migrated between so that its
not an issue. Then the only time this would then come into play is
when you are upgrading the group/cluster of machines to newer hardware
and that shouldn't be an issue as the newer hardware should have more
features in the CPU and not less. This is the way we do it in the
Enterprise hosting company that I work for which has a VM environment
over around 250 hosts runninng around 2500 VMs. We have a few
disparate clusters with different devices using different HW but in
that case we use the CPU masking features of the main proprietary
product that is in use in that particular case.

Cheers,
Peter