Paradox in parallel processing
Hi users of Fluent,
I simulate in 1 node processor and the system load is full. When I simulate in a parallel processing with 2 nodes I notice that the system load for the Fluent process is approximately 45% in each node. Why is this happening? Is this right? I would be grateful if someone could explain to me this paradox! My system configuration is: Parallel processing, 2 nodes, Fluent 6.2.16, socket protocol, Linux 64bit, opteron CPU, 2Gb RAM, Gigabit, case file appr. 700000cells Thanks, Vaggelis |
All times are GMT -4. The time now is 03:34. |