Closed shengyang-3390 closed 2 years ago
I test ib_send_bw for two case on ARM services and IB card. Case1:numactl -C 64 ib_send_bw -d mlx5_0 -a --iters=1000 Case2:numactl -C 64 ib_send_bw -d mlx5_0 --size=1024 --iters=1000
Case1 perfermance data:
1024 1000 0.00 5544.28 5.677340
Case2 perfermance data:
1024 1000 0.00 920.13 0.942218
The performance of the case1 is 6 times of case2 performance. What may cause the performance gap?
I don't notice such degradation on my setup. What about the other message sizes?
Closing the Issue, please reopen if repro
I test ib_send_bw for two case on ARM services and IB card. Case1:numactl -C 64 ib_send_bw -d mlx5_0 -a --iters=1000 Case2:numactl -C 64 ib_send_bw -d mlx5_0 --size=1024 --iters=1000
Case1 perfermance data:
bytes #iterations BW peak[MB/sec] BW average[MB/sec] MsgRate[Mpps]
1024 1000 0.00 5544.28 5.677340
Case2 perfermance data:
bytes #iterations BW peak[MB/sec] BW average[MB/sec] MsgRate[Mpps]
1024 1000 0.00 920.13 0.942218
The performance of the case1 is 6 times of case2 performance. What may cause the performance gap?