RE: fork c/c++ computational problems
"Jason Coutu" <[email protected]>
| Newsgroups | gmane.linux.cluster.openmosix.general |
|---|---|
| Message-ID | <[email protected]> |
Forking is simply having your program continue on from the location of the fork as two processes instead of one. What is usually done is that one of the two processes (the parent[creator] and the child[created])does somthing slightly differnt. If the two processes need to syncronize or communicate you will need to handle that seperately. MPI is one of the options for comminication, there are also semaphores, sockets and named pipes. Jason > -----Original Message----- > From: openmosix-general-admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org > [mailto:openmosix-general-admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org]On Behalf Of Amin > Patel > Sent: Wednesday, December 28, 2005 6:45 AM > To: openmosix-general-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org > Subject: [openMosix-general] fork c/c++ computational problems > > > Hi, I'm an Engineering student and got intrested in openmosix for > computational problems. I got an openmosix cluster of 3 nodes to > function and talk to each other. Now here is my observation, if I write > "normal" c/c++ code the processes "migrate" or get "distributed" only > when I run multiple instances of the program. That is not a bad thing, > as such a functionality would have its own uses. Now my purpose or ideal > goal is to be able to write programs or computational code that I only > run once, but the nature of the calculations would force openmosix to > distribute the load amongst the nodes. > > As i understand 'fork'-ing is a solution to this problem, but I'm unable > to find good documentation or programming code that utilizes such a > solution. In addition I'm finding a lot on MPI written code, however, I > do not feel comfortable with MPI+openmosix idea as of yet (just started > with this endeavour, given my recent entry into linux). Does anyone know > of any tutorials or resources which involves computational problems, > that utilize 'fork', and are not meant to be run as multiple instances > but only once instance, which ends up distributing the load, without MPI. > > Cheers... > amin > > > > ------------------------------------------------------- > This SF.net email is sponsored by: Splunk Inc. Do you grep > through log files > for problems? Stop! Download the new AJAX search engine that makes > searching your log files as easy as surfing the web. DOWNLOAD SPLUNK! > http://ads.osdn.com/?ad_id=7637&alloc_id=16865&op=click > _______________________________________________ > openMosix-general mailing list > openMosix-general-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org > https://lists.sourceforge.net/lists/listinfo/openmosix-general ------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Do you grep through log files for problems? Stop! Download the new AJAX search engine that makes searching your log files as easy as surfing the web. DOWNLOAD SPLUNK! http://ads.osdn.com/?ad_id=7637&alloc_id=16865&op=click