abcdabcd987 / libfabric-efa-demoLinks
☆73Updated last year
Alternatives and similar repositories for libfabric-efa-demo
Users that are interested in libfabric-efa-demo are comparing it to the libraries listed below
Sorting:
- NVSHMEM‑Tutorial: Build a DeepEP‑like GPU Buffer☆151Updated 3 months ago
- DeepXTrace is a lightweight tool for precisely diagnosing slow ranks in DeepEP-based environments.☆83Updated 2 weeks ago
- A lightweight design for computation-communication overlap.☆207Updated 2 weeks ago
- Stateful LLM Serving☆92Updated 9 months ago
- ☆47Updated last year
- Thunder Research Group's Collective Communication Library☆46Updated 6 months ago
- ☆81Updated 2 months ago
- A NCCL extension library, designed to efficiently offload GPU memory allocated by the NCCL communication library.☆77Updated 3 weeks ago
- Microsoft Collective Communication Library☆66Updated last year
- High performance Transformer implementation in C++.☆148Updated 11 months ago
- ☆337Updated this week
- A prefill & decode disaggregated LLM serving framework with shared GPU memory and fine-grained compute isolation.☆123Updated 2 weeks ago
- Aims to implement dual-port and multi-qp solutions in deepEP ibrc transport☆73Updated 7 months ago
- ☆72Updated 3 months ago
- High-performance distributed data shuffling (all-to-all) library for MoE training and inference☆102Updated last week
- Automated Parallelization System and Infrastructure for Multiple Ecosystems☆82Updated last year
- NEO is a LLM inference engine built to save the GPU memory crisis by CPU offloading☆77Updated 6 months ago
- Efficient Compute-Communication Overlap for Distributed LLM Inference☆67Updated 2 months ago
- ☆49Updated 8 months ago
- Paella: Low-latency Model Serving with Virtualized GPU Scheduling☆66Updated last year
- ☆70Updated 2 months ago
- NVIDIA NVSHMEM is a parallel programming interface for NVIDIA GPUs based on OpenSHMEM. NVSHMEM can significantly reduce multi-process com…☆433Updated last week
- DeeperGEMM: crazy optimized version☆74Updated 8 months ago
- ☆153Updated last year
- Chimera: bidirectional pipeline parallelism for efficiently training large-scale models.☆69Updated 9 months ago
- Artifact of OSDI '24 paper, ”Llumnix: Dynamic Scheduling for Large Language Model Serving“☆64Updated last year
- DLSlime: Flexible & Efficient Heterogeneous Transfer Toolkit☆91Updated this week
- torchcomms: a modern PyTorch communications API☆315Updated last week
- nnScaler: Compiling DNN models for Parallel Training☆122Updated 3 months ago
- FlexFlow Serve: Low-Latency, High-Performance LLM Serving☆66Updated 3 months ago