Florian
2008-09-23 10:36:18 UTC
Hi,
I am currently trying to implement a parallel version for a program
that requires a lot of memory (>6GB). The problem is that the cluster
I have available has 40 nodes with 2 dual core processors that share
4GB memory.
My plan is to distribute the memory over several nodes.
Could anyone tell me how I can find out on which node which processor
is so that I can use the 4 processors of each node most effectively? I
want to use each processor for problems that require only data
allocated on their own node.
Thanks a lot,
Florian
I am currently trying to implement a parallel version for a program
that requires a lot of memory (>6GB). The problem is that the cluster
I have available has 40 nodes with 2 dual core processors that share
4GB memory.
My plan is to distribute the memory over several nodes.
Could anyone tell me how I can find out on which node which processor
is so that I can use the 4 processors of each node most effectively? I
want to use each processor for problems that require only data
allocated on their own node.
Thanks a lot,
Florian