[MPICH] Stopping processors?

[Michaela Heyer]

Yeah I know that but sure I'd still have to have at least one, possibly more 
MPI_Iprobe statements in each algorithm, no?

On Thursday 26 April 2007 15:01, Rajeev Thakur wrote:
> You can do a single MPI_Iprobe with a wildcard src and tag, MPI_ANY_SOURCE,
> MPI_ANY_TAG. It will match a stop message from anyone.
>
> Rajeev
>
> > -----Original Message-----
> > From: owner-mpich-discuss at mcs.anl.gov
> > [mailto:owner-mpich-discuss at mcs.anl.gov] On Behalf Of Michaela Heyer
> > Sent: Thursday, April 26, 2007 4:10 AM
> > To: mpich-discuss at mcs.anl.gov
> > Subject: Re: [MPICH] Stopping processors?
> >
> > Hi Guys,
> > Thanks for the replies!
> >
> > Rajeev, what you suggested was my initial idea also. The
> > problem with this is
> > that I have about 200 different algorithms so the way I see
> > it I'd have to
> > add MPI_Iprobe (actually I was gonna use MPI_Test) to all the
> > algorithms and
> > possible call it even more than once to really make sure I
> > get stopped quite
> > quickly...which just doesn't seem feasible.
> >
> > Darius, I like the idea with the threads! I haven't really
> > worked much with
> > threads, so I will have to look into it but I think it might work...
> >
> > Michaela
> >
> > On Tuesday 24 April 2007 16:47, Darius Buntinas wrote:
> > > One idea would be to have each worker process spawn a "computation"
> > > thread which actually does the computation.  The "main"
> >
> > thread of the
> >
> > > process would issue a blocking receive waiting for a "done" message.
> > >
> > > When a computation thread at some process finishes, it
> >
> > sends messages to
> >
> > > the other processes and to itself.  When a main thread
> >
> > receives a done
> >
> > > message, it checks whether it received the message from its own
> > > computation thread.  If so, it does a pthread_join() and reads the
> > > result, otherwise it does a pthread_cancel().
> > >
> > > At this point the main thread can wait for a message to
> >
> > start the next
> >
> > > computation.
> > >
> > >
> > > If you have a lot of processes having one process send a
> >
> > done message to
> >
> > > every other process isn't scalable.  You could optimize in
> >
> > that case by
> >
> > > having the processes forward the message in a tree.
> > >
> > > -d
> >
> > On Tuesday 24 April 2007 16:59, Rajeev Thakur wrote:
> > > You could have all processes post an MPI_Irecv and then
> >
> > periodically call
> >
> > > MPI_Iprobe to see if there is an incoming message. The process that
> > > finishes first sends others a stop message.
> > >
> > > Rajeev
> > >
> > > > -----Original Message-----
> > > > From: owner-mpich-discuss at mcs.anl.gov
> > > > [mailto:owner-mpich-discuss at mcs.anl.gov] On Behalf Of
> >
> > Michaela Heyer
> >
> > > > Sent: Tuesday, April 24, 2007 9:21 AM
> > > > To: mpich-discuss at mcs.anl.gov
> > > > Subject: [MPICH] Stopping processors?
> > > >
> > > > Hi,
> > > > I'm hoping someone can help me with this...
> > > > Essentially I'm looking for a way to tell one or more
> > > > processors to stop
> > > > whatever they are currently doing and move on to
> >
> > something else. The
> >
> > > > situation is as follows: Let's say we have n processors all
> > > > working on
> > > > different algorithms. I only really need the result of
> >
> > one of these
> >
> > > > algorithms so whichever processor finishes first should tell
> > > > all the other
> > > > ones to stop so they can all move on to the next problem.
> > > > It's a bit like a
> > > > race...and speed is vital! (That's why I can't really wait
> > > > for all processors
> > > > to finish their algorithms)
> > > > At the moment I'm using MPI_Abort() to shutdown the whole
> > > > process. This works
> > > > fine and does exactly what I need but the problem is that it
> > > > is very very
> > > > slow as it shuts down and restarts all the processors. So
> > > > what I'm looking
> > > > for is something like a "milder" version of MPI_Abort() i.e.
> > > > stop everything
> > > > but don't shutdown the processors.
> > > > I have been looking but can't find anything sutitable so I'm
> > > > starting to think
> > > > that maybe it's impossible? It would be great if you could
> > > > prove me wrong :-)
> > > > Thanks,
> > > > Michaela