NVIDIA/nvshmem
NVIDIA NVSHMEM is a parallel programming interface for NVIDIA GPUs based on OpenSHMEM. NVSHMEM can significantly reduce multi-process communication and coordination overheads by allowing programmers to perform one-sided communication from within CUDA kernels and on CUDA streams.
This is a parallel programming interface for NVIDIA GPUs that helps developers of high-performance computing applications to build programs that communicate efficiently across multiple GPUs. It allows direct, one-sided communication between GPUs, reducing the overhead typically associated with data sharing in large-scale simulations or machine learning models. HPC developers or researchers working with GPU clusters will find this useful for optimizing their multi-GPU applications.
477 stars.
Use this if you are developing high-performance applications on NVIDIA GPU clusters and need to optimize communication between GPUs for speed and efficiency.
Not ideal if you are working with single-GPU applications or do not require fine-grained control over GPU communication.
Stars
477
Forks
65
Language
C++
License
—
Category
Last pushed
Mar 10, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/NVIDIA/nvshmem"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related frameworks
tensorflow/tensorflow
An Open Source Machine Learning Framework for Everyone
microsoft/tensorwatch
Debugging, monitoring and visualization for Python Machine Learning and Data Science
KomputeProject/kompute
General purpose GPU compute framework built on Vulkan to support 1000s of cross vendor graphics...
hailo-ai/hailort-drivers
The Hailo PCIe driver is required for interacting with a Hailo device over the PCIe interface
hidet-org/hidet
An open-source efficient deep learning framework/compiler, written in python.