Proposal: Fedora at Home Project

[Jeff Spaleta]

On Mon, Mar 10, 2008 at 1:16 PM, Bryan Che <bche at redhat.com> wrote:
>  I think we should start with Fedora infrastructure needs and then expand
>  to other computational problems.  Getting it right first for Fedora will
>  give us a good foundation for adding other projects.

I'm not sure I have a good grasp on where we could reasonably employ
grids that benefits any ongoing internal work. My understanding is
that we aren't straining our existing buildhost infrastructure, so its
not clear there is a need here to farm out builds and then send the
binaries back over the wire.  Do we have an internal need identified
that could leverage a grid of cpu time?  I'm not even sure I remember
seeing anyone complain yet.  Maybe there's a real interesting case for
this for secondary arches...if we had cross compiler tools so idle
client pc time could be use to help build arm/sparc/etc packages. I'm
just not sure we've got an interesting internal need for idle cpu
cycles.

I see this being more interesting as a way to have Fedora as a
community interact with outside computing efforts. Off the top of my
head, let's say we had this infrastructure up and running. Could we
reach out to the blender community and work with them by offering the
Fedora grid to help render scenes of another community funded animated
film?

-jef"I hope Greg is reading this"spaleta