ALLAH
Dear Michael,
our OSCAR cluster has 18 nodes with 8 cpus per node so
it has 144 cpus(xeon e5335*2)
I am running pwscf ((it is open source (pwscf.org)) by
mpich2.But there is a problem when i improve unmber of
cpus per nodes in my calculation, the program stop with no
error showing. I am sure that not my ram overloade. because when
i run a pwscf program in one of my nodes(as shared memory
machine with 8 cpus) it works completely.
Do u have any opinion a bout this problem?(i have a gigabit
ethernet.)
regards siamak jadidi
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
-------------------------------------------------------------------------
This SF.net email is sponsored by the 2008 JavaOne(SM) Conference
Don't miss this year's exciting event. There's still time to save $100.
Use priority code J8TL2D2.
http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
_______________________________________________
Oscar-users mailing list
Oscar-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/oscar-users
