Opinions on packaging ATLAS (for the x86 architecture)

Jeff Spaleta jspaleta at gmail.com
Wed Oct 7 16:21:12 UTC 2009


On Wed, Oct 7, 2009 at 7:29 AM, Bill Nottingham <notting at redhat.com> wrote:
> Surely the way to do this is to know what your workload is doing,
> and not do live migration to random hardware?

I think random hardware is going to be exactly what you will see a lot
of scientific research appliances being thrown at.  There are groups
interested in offering off the shelf appliances for some complex
computational codes meant to be run on private/academic/national lab
infrastructure to make it easier for scientific users to run the codes
correctly.  It's pretty pedantic stuff, and you can really screw up
the build configuration when you are trying to build your own
instances (DYI kernel compiles can be refreshingly less complicated
sometimes)  Live migrating appliances across those boundaries will
undoubtedly be desired.  National Lab run "clouds" are going to be
well defined hardware targets.... but academic clouds are going to be
all over the place.

-jef




More information about the fedora-devel-list mailing list