Re: process communication
Tom Davis <[email protected]>
| Newsgroups | gmane.linux.cluster.openmosix.general |
|---|---|
| Message-ID | <[email protected]> |
I'm by no means any kind of authority on this, but openMosix shouldn't have a problem with migrating Sys V IPC Message Queues - doesn't need any kind of patching as far as I'm aware. Did a University project on it a year ago and it certainly seemed to be migrating our processes. :-) One thing to look at though - how much are the queues being used? Every time a system call (e.g. to use a message queue) is made, it is migrated back to the home-node. If many are being made, openMosix makes the decision that it is quicker for both processes to execute on the home node. If this is the case, then pipe or whatever, your process will not leave the home node. Anyway, as I say - I'm by no means an authority and haven't used openMosix for about six months! So am willing to be corrected here. ;-) HTH, Tom Adrian wrote: > Hello, > > I want to write a c++-application for openMosix with some kind of process > communication. First I tried it with shared-memory, now with message queues. > With both methods openMosix did not migrate the processes. I've seen that > there's a kernel patch to enable shared memory, I will try it. But I just > wonder, whether there's a simple way to establish process communication > which works under openMosix > > BTW, I could nowhere find a list of c/c++ system-calls that prevent > openMosix from migrating the process. Does anyone have one or know where to > find one? > > Thanks for your help! > Adrian > > > > ------------------------------------------------------- This SF.Net email is sponsored by the JBoss Inc. Get Certified Today * Register for a JBoss Training Course Free Certification Exam for All Training Attendees Through End of 2005 Visit http://www.jboss.com/services/certification for more information