microsoft / ark
A GPU-driven system framework for scalable AI applications
☆109Updated last month
Related projects ⓘ
Alternatives and complementary repositories for ark
- MSCCL++: A GPU-driven communication stack for scalable AI applications☆250Updated this week
- An experimental parallel training platform☆52Updated 7 months ago
- NCCL Profiling Kit☆112Updated 4 months ago
- Microsoft Collective Communication Library☆53Updated last month
- Synthesizer for optimal collective communication algorithms☆99Updated 7 months ago
- Paella: Low-latency Model Serving with Virtualized GPU Scheduling☆57Updated 6 months ago
- Artifact of OSDI '24 paper, ”Llumnix: Dynamic Scheduling for Large Language Model Serving“☆57Updated 5 months ago
- A Python library transfers PyTorch tensors between CPU and NVMe☆98Updated last week
- High performance Transformer implementation in C++.☆82Updated 2 months ago
- ☆73Updated last year
- Microsoft Collective Communication Library☆323Updated last year
- Dynamic Memory Management for Serving LLMs without PagedAttention☆238Updated last week
- Automated Parallelization System and Infrastructure for Multiple Ecosystems☆75Updated this week
- An experimental CPU backend for Triton☆56Updated last week
- PArametrized Recommendation and Ai Model benchmark is a repository for development of numerous uBenchmarks as well as end to end nets for…☆124Updated this week
- ☆45Updated 2 weeks ago
- An experimental CPU backend for Triton (https//github.com/openai/triton)☆35Updated 6 months ago
- [OSDI'24] Serving LLM-based Applications Efficiently with Semantic Variable☆114Updated 2 months ago
- A validation and profiling tool for AI infrastructure☆273Updated this week
- Unified Collective Communication Library☆207Updated last week
- ☆131Updated 3 months ago
- A fast communication-overlapping library for tensor parallelism on GPUs.☆224Updated 3 weeks ago
- PET: Optimizing Tensor Programs with Partially Equivalent Transformations and Automated Corrections☆114Updated 2 years ago
- TiledCUDA is a highly efficient kernel template library designed to elevate CUDA C’s level of abstraction for processing tiles.☆154Updated this week
- Intel® Extension for DeepSpeed* is an extension to DeepSpeed that brings feature support with SYCL kernels on Intel GPU(XPU) device. Note…☆57Updated 2 months ago
- ☆33Updated 2 months ago
- Shared Middle-Layer for Triton Compilation☆191Updated this week
- ☆55Updated 5 months ago
- ☆46Updated 5 months ago
- An IR for efficiently simulating distributed ML computation.☆25Updated 10 months ago