Search results

Asymmetric bandwidth in bidirectional Host-Device transfers: H2D drops to ~11 GB/s while D2H maintains ~25 GB/s

Status: Open.

#53 In NVIDIA/nvbandwidth;

· chanpapa opened on Nov 21, 2025

Could you please provide reference results?

Status: Open.

#50 In NVIDIA/nvbandwidth;

· rsdmse opened on Nov 3, 2025

performance of device_local_copy does not meet expectations on NVIDIA H20-3e

Status: Open.

#48 In NVIDIA/nvbandwidth;

· yoqiu-amd opened on Sep 17, 2025

multinode run fails with [CUDA_ERROR_NOT_PERMITTED] error in cuMemCreate() call

Status: Open.

#46 In NVIDIA/nvbandwidth;

· robertwenquan opened on Aug 27, 2025

multinode benchmarks fail to start with error - [CUDA_ERROR_INVALID_VALUE] invalid argument in expression cuMemSetAccess

Status: Open.

#44 In NVIDIA/nvbandwidth;

· chtcvl opened on Jul 29, 2025

Wrong Documentation message on giving buffer size

Status: Open.

#43 In NVIDIA/nvbandwidth;

· alokprasad opened on Jul 23, 2025

Performance drop in host_to_device_memcpy_sm with large buffers

Status: Open.

#42 In NVIDIA/nvbandwidth;

· yuyuankang opened on Jul 5, 2025

SM memcpy performance concerns on GH200

Status: Open.

#41 In NVIDIA/nvbandwidth;

· rajesh-s opened on May 19, 2025

why is no all-to-one with bidirectional test

Status: Open.

#38 In NVIDIA/nvbandwidth;

· houqi opened on Apr 21, 2025

which metric is spec bandwidth for a gpu

Status: Open.

#36 In NVIDIA/nvbandwidth;

· ltm920716 opened on Apr 3, 2025

./nvbandwidth -t device_to_device_memcpy_read_ce failed due to unknown issue

Status: Open.

#34 In NVIDIA/nvbandwidth;

· plorrrrrrr opened on Feb 17, 2025

Support for Jetson GPUs (Jetpack 6.0+, Orin)

Status: Open.

#28 In NVIDIA/nvbandwidth;

· dicta opened on Dec 16, 2024