And for a node, is there allso a method which isn't LAM-specific? Which can
be used with all MPI-2 implementations?
(can't you use allso OPEN_PORT, COMM_ACCEPT/CONNECT on the "new" node to
connect it with a node which "allready knows what to do" ?
Or are there any other methods?
Thanks.
On 8/9/05, Jeff Squyres <jsquyres_at_[hidden]> wrote:
>
> Are you asking about a *node* joining the universe, or a *process*
> joining a computation? Or both?
>
> For a node to join, you need to use the lamgrow command.
>
> For a process to join, yes, you can to use the OPEN_PORT,
> COMM_ACCEPT/CONNECT calls. Check the MPI-2 standard in the dynamic
> chapter; there are code examples in there.
>
>
>
> On Aug 8, 2005, at 10:58 AM, Jim Lasc wrote:
>
> > Hi, I searched a lot on node insertion for dynamic clusters.
> > All I could find was this post from this mailinglist:
> > ------
> >
> > This only "sorta" supported. Note that lamgrow / lamshrink really only
> > expand / shrink the LAM *univerise*, not a running application. You'll
> > need to add special stuff into an application to make it aware that
> > the
> > universe has grown or shunk.
> >
> > For example, you may want to have one "master" process that
> > MPI_COMM_SPAWN's a bunch of workers and dynamically gives them work to
> > do. As a new worker becomes available, the master can be notified
> > somehow (perhaps via some mechanism outside of MPI, such as a socket,
> > pipe, file, or other IPC mechanism) and it can spawn a new worker
> > there. Similarly, when a node becomes unavailable, the master can be
> > told, and it can send a "please shut yourself down" message to the
> > worker.
> > -----
> > I have some questions about this:
> > Can someone give me example code for a node which joins the cluster
> > and starts calculating. I tried to make it myself, but apparently it's
> > too hard for me to do so.
> >
> > I guess I should use MPI_Open_Port , MPI_Comm_accept and
> > MPI_Comm_spawn, but didn't managed into getting something to work
> > (and my attempts are to pathetic to post)...
> >
> > Any help is really appreciated.
> > Jim Lascov.
> > _______________________________________________
> > This list is archived at http://www.lam-mpi.org/MailArchives/lam/
>
> --
> {+} Jeff Squyres
> {+} jsquyres_at_[hidden]
> {+} http://www.lam-mpi.org/
>
>
> _______________________________________________
> This list is archived at http://www.lam-mpi.org/MailArchives/lam/
>
|
