MLNX_OFED GPUDirect RDMA
The latest advancement in GPU-GPU communications is GPUDirect RDMA. This technology provides a direct P2P (Peer-to-Peer) data path between the GPU Memory directly to/from the NVIDIA networking adapter devices. This provides a significant decrease in GPU-GPU communication latency and completely offloads the CPU, removing it from all GPU-GPU communications across the network. GPU Direct leverages PeerDirect RDMA and PeerDirect ASYNC capabilities of the NVIDIA network adapters.
- Avoid unnecessary system memory copies and CPU overhead by copying data directly to/from pinned GPU memory
- Peer-To-Peer Transfers Between GPU device and NVIDIA RDMA-based networking devices
- Use high-speed DMA transfers to copy data between P2P devices
- Eliminate CPU bandwidth and latency bottlenecks using direct memory access (DMA)
- With GPUDirect RDMA, GPU memory can be used for Remote Direct Memory Access (RDMA) resulting in more efficient applications
- Boost Message Passing Interface (MPI) Applications with zero-copy support