`host_to_device_memcpy_sm` low than `device_to_host_memcpy_sm`

[ywxc1997]

Hi!
When we tested host_to-vice_cemcpy_sm and Device_to-host_cemcpy_sm separately on the H100 cluster, we obtained two completely different values

Running host_to_device_memcpy_sm.
memcpy SM CPU(row) -> GPU(column) bandwidth (GB/s)
0 1 2 3 4 5 6 7
0 35.19 35.25 35.30 35.03 35.25 35.32 35.39 35.06

Running device_to_host_memcpy_sm.
memcpy SM CPU(row) <- GPU(column) bandwidth (GB/s)
0 1 2 3 4 5 6 7
0 52.77 52.77 52.77 52.78 52.76 52.77 52.78 52.77

Actually, they should be close values.
What could be causing this?

[jessicaSGH]

I got the similar test results with H20 and H800

[zhangvia]

hey，how did you build the nvbandwidth on H100? i encounter build failure on rtx4090, it says sm89 not supported

[eomhyeonpil]

@ywxc1997
Could you let me know if you got your questions answered?
if the reason as below?