Notes aout NVLink and NVSwitch。

NVLink is a wire-based serial multi-lane near-range communications link developed by Nvidia. Unlike PCI Express, a device can consist of multiple NVLinks, and devices use mesh networking to communicate instead of a central hub.

2. NVSwitch


In the above figure, GPU to GPU memory transfers via NVLink are at most two hops away – a memory request may have to be routed through the NVLink controllers on two GPUs. For example, GPU 0 may need data in GPU 5’s memory, it needs two hops (such as:GPU 0 -> GPU 1 ->GPU5). Each NVLink controller has a memory access latency, so each memory access latency multiplies via the number of hops is the total latency.

NVSwitch存在的作用是避免GPU和GPU之间的通信会存在多跳。

参考资料:

  1. https://www.cnblogs.com/kongchung/p/12945019.html
  2. https://www.nvidia.com/en-us/data-center/nvlink/
  3. https://en.wikichip.org/wiki/nvidia/nvlink
  4. https://en.wikichip.org/wiki/nvidia/nvswitch
  5. https://en.wikipedia.org/wiki/NVLink
  6. https://www.nextplatform.com/2018/04/13/building-bigger-faster-gpu-clusters-using-nvswitches/