On 11/23/2010 06:47 AM, Barrett, Brian W wrote:
Short answer: we need the "extra" decrement at the end of MPI init.
Long answer: Ok, so I was somewhat wrong :).
.... (surprised this didn't show up in testing).
Confirmed with our basic pingpong test:
vayu2:~/MPI > mpirun -n 2 ./pp142 | head -4
| Best | Average | Worst
| Streaming
n (test/loop) | usec MB/s | usec MB/s | usec
MB/s | ( average, worst)
1 ( 64/1100) | 0.3 23.56 | 0.4 19.69 | 0.6
14.24 | ( 0.4, 0.5)
2 ( 64/1100) | 0.3 46.93 | 0.4 39.43 | 0.5
29.36 | ( 0.4, 0.5)
vayu2:~/MPI > mpirun -n 2 ./pp143 | head -4
| Best | Average | Worst
| Streaming
n (test/loop) | usec MB/s | usec MB/s | usec
MB/s | ( average, worst)
1 ( 64/1100) | 0.5 16.84 | 0.6 13.72 | 0.8
9.93 | ( 0.6, 0.8)
2 ( 64/1100) | 0.5 32.93 | 0.6 26.97 | 0.8
20.01 | ( 0.6, 0.8)
David
