> It would be easy for people to just say "Well, those SGI engineers are
> stupid". Heck, I could say it, I didn't work on the migration stuff,
> I thought it was a bad idea before they started working on it, so it
> isn't like I have any skin in the idea, quite the opposite. But the
> SGI engineers are top notch, at least in this area. I challenge the
> minds out there to come up with a migration policy that actually
> improves performance under any realistic workload, not some toy
> benchmark. Show me a process migration system that makes TPC-C
> run better. Or fortran jobs. Or make. Or web. Or NFS. Anything
> that customers will pay money to get.
Maybe I'm just missing something, but is performance considered to be the
ONLY potential advantage of inter-machine process migration?
How about uses for high availability, for e.g. ? This would probably
require some kind of checkpointing depending on what kind of failures
we're trying to recover from, or there may be a better way to get state
info migrated as well.
No?
---------------------------------------------------------------|-----|--
Louis Mandelstam Tel +27 83 227-0712 Symphony /|\ /|\
Linux systems integration http://sr.co.za Research { } { }
Johannesburg, South Africa mailto:louis@sr.co.za (Pty)Ltd {___} {___}