mirror of
https://github.com/NVIDIA/TensorRT-LLM.git
synced 2026-01-14 06:27:45 +08:00
fd4311e6a3
Because we have encountered some perf regression due to using a one-shot kernel instead of NCCL on A100/H100, it will be beneficial if we can have a solid benchmarking of allreduce Op and analyze the data collected from it. Implemented new AllreduceOp heuristic: - Added Linear programming-based heuristic implementation. - Added LUT-based heuristic implementation and corresponding code generation script. AllreduceOp minor fixing: - Fixed a minor issue in AllreduceOp, that the strategy can not be overridden when ONESHOT or TWOSHOT is set. - Fixed a minor TWOSHOT kernel perf issue. - Cleaned up Dispatching code in AllReduceOp. This PR will fix the perf gaps reported in: https://nvbugspro.nvidia.com/bug/5517023 For Deepseek-R1, it shows a performance gain of about 3-4% in concurrency levels of 256 and 512. Signed-off-by: Yukun He <23156053+hyukn@users.noreply.github.com> |
||
|---|---|---|
| .. | ||
| allReduceFusionKernels.cu | ||
| allReduceFusionKernels.h | ||
| allReduceWorkspace.cu | ||
| allReduceWorkspace.h | ||
| customLowPrecisionAllReduceKernels.cu | ||
| customLowPrecisionAllReduceKernels.h | ||
| mnnvlTwoShotAllreduceKernels.cu | ||
| mnnvlTwoShotAllreduceKernels.h | ||
| moeAllReduceFusionKernels.cu | ||
| moeAllReduceFusionKernels.h | ||