OK, but finish the argument. If all the other CPUs are busy doing their
larger blocks of rays, how does migration help me in this case?
: > All the parallel computations I've seen statically load balance at
: > initialization time and then never move.
:
: Because people use them for customised specific jobs, partly because thats
: the most efficient way to do it in CPU terms (and Im not arguing that issue
: - DSM is less efficient)
So are you just giving up on the efficiency issue? It's OK to throw
performance away because you are using DSM?
: > I agree that there are other types of parallel loads, but they
: > tend to be more of the time sharing sort. I don't think anyone
: > is suggesting that migrating vi or sendmail is a good idea. Right?
:
: What about big background computations spread across a network of machines
: also used for desktop timesharing - Condor does exactly this. People login
: to a desktop and the fortran monster quietly sidles off to another work
: station.
Sure, for that sort of application, migration is great. As long as the
process of "quietly sliding off" doesn't involve 300MB of data that the
fortran job was chewing on.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu