Does anyone have host-to-device and device-to-host bandwidth performance numbers for PCIe 2.0-based GPUs?
RHsu
1
[url=“The Official NVIDIA Forums | NVIDIA”]http://forums.nvidia.com/index.php?showtop...ndpost&p=320460[/url]
And the post above it.
Related topics
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| Host to Device Bandwidth and PCEe 2.0 - not getting what I should! | 4 | 2806 | February 18, 2009 | |
| CUDA on PCI 1.0 | 3 | 4134 | December 15, 2009 | |
| Host to Device Memroy Bandwidth | 18 | 8195 | September 12, 2008 | |
| what is 'bandwidth' from where to where | 2 | 1283 | January 4, 2009 | |
| PCIe bandwidth is asymmetrical between Host to Device and Device to Host | 0 | 298 | January 2, 2025 | |
| Is my bandwidth calculation right? bandwidth | 3 | 1553 | November 13, 2009 | |
| C1060 Bandwidths | 3 | 1943 | June 11, 2009 | |
| CudaMemcpy() speed/bandwidth For host to device | 5 | 10163 | June 30, 2009 | |
| bandwidth question | 2 | 2169 | August 15, 2008 | |
| How to measure bandwidth from pinned host memory to device memory on aws A100(p4d.24xlarge)? | 1 | 1056 | September 7, 2022 |