Hi all,
I have a question which puzzles me, about running LAM.
I have a 32-bit Dual Intel Xeon machine, and an AMD 64-bit Dual Opteron machine.
Both of them are running Solaris 10, Build 72.
I want to know whether there is a difference in performance between running LAM across two 32-bit machines, or across one 32, and one 64-bit machine.
I use HPL 1.0a to try to investigate this issue, but the performance ratings are the same, for both conditions.
Or do I need both nodes to be 64-bit to make use of 64-bit performance?
In my boot schema file, I did specify "cpu=2" for the both machines. Does this affect anything? Or should I just leave it out?
I used the following command to run the benchmark program, with the 'C' option.
mpirun C xhpl
The HPL program running from the 32-bit node is compiled with the 32-bit LAM, and the same program on the 64-bit node compiled with the 64-bit LAM.
I want to know the differences, so I can deduce whether LAM is really properly compiled as a 64-bit application.
Quite a few questions here... Will really appreciate any feedback. Please tell me if you need more information on the operating environment, or compile issues.
Thank you for your time.
Regards,
Juncheng
|
