GPU Direct is addressed in FAQ ( DGX Spark / GB10 FAQ )
Related topics
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| Potential NCCL bug in topology discovery in NCCL2.1.15 | 0 | 1289 | March 16, 2018 | |
| GPUDirect RDMA support with CUDA 5 | 19 | 9362 | May 28, 2013 | |
| Small BUG in deviceQuery.cpp | 0 | 870 | January 5, 2014 | |
| GPUDirect RDMA for Ethernet Cards | 1 | 818 | September 20, 2013 | |
| GPUDirect RDMA performance | 2 | 2231 | March 26, 2013 | |
| How can I tell whether NCCL is using PCIe or IB network interface while doing AllReduce? | 0 | 784 | March 6, 2020 | |
| Issues with Horovod multi-node connection over InfiniBand + GPUDirect RDMA | 1 | 1044 | January 2, 2020 | |
| Fast Multi-GPU collectives with NCCL | 14 | 1210 | May 11, 2018 | |
| Trying to get GPUdirect RDMA working. | 2 | 1671 | April 10, 2014 | |
| Partial fail of peer access in 8 Volta GPU instance (p3.16xlarge) on AWS -> huge slowdown | 32 | 3820 | March 10, 2018 |