muriloboratto / NCCL
Sample examples of how to call collective operation functions on multi-GPU environments. A simple example of using broadcast, reduce, allGather, reduceScatter and sendRecv operations.
☆32Updated last year
Alternatives and similar repositories for NCCL:
Users that are interested in NCCL are comparing it to the libraries listed below
- ☆90Updated 3 weeks ago
- Matrix Multiply-Accumulate with CUDA and WMMA( Tensor Core)☆127Updated 4 years ago
- Several optimization methods of half-precision general matrix vector multiplication (HGEMV) using CUDA core.☆58Updated 6 months ago
- Optimize GEMM with tensorcore step by step☆24Updated last year
- We invite you to visit and follow our new repository at https://github.com/microsoft/TileFusion. TiledCUDA is a highly efficient kernel …☆179Updated 2 months ago
- ⚡️Write HGEMM from scratch using Tensor Cores with WMMA, MMA and CuTe API, Achieve Peak⚡️ Performance.☆63Updated this week
- TileFusion is an experimental C++ macro kernel template library that elevates the abstraction level in CUDA C for tile processing.☆74Updated this week
- Standalone Flash Attention v2 kernel without libtorch dependency☆108Updated 6 months ago
- ☆112Updated 3 months ago
- Performance of the C++ interface of flash attention and flash attention v2 in large language model (LLM) inference scenarios.☆35Updated last month
- ☆60Updated 3 months ago
- ☆49Updated 5 years ago
- An extension library of WMMA API (Tensor Core API)☆93Updated 8 months ago
- 使用 cutlass 仓库在 ada 架构上实现 fp8 的 flash attention☆60Updated 7 months ago
- Examples of CUDA implementations by Cutlass CuTe☆148Updated last month
- CUTLASS and CuTe Examples☆43Updated 2 months ago
- ☆92Updated 11 months ago
- ☆48Updated 2 months ago
- ☆73Updated 4 months ago
- CUDA Matrix Multiplication Optimization☆177Updated 8 months ago
- DeeperGEMM: crazy optimized version☆63Updated 2 weeks ago
- Benchmark code for the "Online normalizer calculation for softmax" paper☆85Updated 6 years ago
- A standalone GEMM kernel for fp16 activation and quantized weight, extracted from FasterTransformer☆90Updated last month
- Implementation of TSM2L and TSM2R -- High-Performance Tall-and-Skinny Matrix-Matrix Multiplication Algorithms for CUDA☆32Updated 4 years ago
- ☆21Updated last month
- ☆45Updated this week
- study of Ampere' Sparse Matmul☆17Updated 4 years ago
- FlexFlow Serve: Low-Latency, High-Performance LLM Serving☆32Updated this week
- Matrix multiplication on GPUs for matrices stored on a CPU. Similar to cublasXt, but ported to both NVIDIA and AMD GPUs.☆31Updated 3 months ago
- ☆15Updated 5 years ago