NVSHMEM

• Added TMA support to NVLink put and get operations, improving single-thread operation performance and enabling non-blocking put operations over NVLink.
• Added GPUNetIO transport providing both kernel-initiated (GDAKI) and proxy-path communication with multi-NIC support and enabling advanced NIC features by using the DOCA SDK.
• Added nvshmemx_flush APIs to provide source-buffer reusability without guaranteeing remote visibility.
• Improved libfabric transport performance for EFA environments.​
• Added NUMA-aware CPU affinity pinning controlled via NVSHMEM_CPU_AFFINITY and NIC assignment controlled via NVSHMEM_NETDEVS_POLICY.
• Added floating-point atomic add/fetch_add APIs with P2P and proxy-backed IBRC support.
• Added experimental logical endpoint/CFT handle support for fabric-PTX unicast communication.
• Switched to C++17 as minimum required C++ version.
• Changed licensing to Apache-2.0.

• Updated Numbast integration and dependency handling for newer Python/Numba-CUDA combinations, including Python 3.14 compatibility.
• Removed hardcoded CUDA 13 build requirement.
• Updated CuTe DSL RMA tensor tests to use Torch-backed tensors with DLPack conversion.​
• Fixed CuTe and Numba device collective generation and small-team handling, including reducescatter cooperative-launch bindings.​

• Combines the memory of multiple GPUs into a partitioned global address space that’s accessed through NVSHMEM APIs
• Includes a low-overhead, in-kernel communication API for use by GPU threads
• Includes stream-based and CPU-initiated communication APIs
• Supports x86 and Arm processors.
• Is interoperable with MPI and other OpenSHMEM implementations

Increase Performance

Convolution is a compute-intensive kernel that’s used in a wide variety of applications, including image processing, machine learning, and scientific computing. Spatial parallelization decomposes the domain into sub-partitions that are distributed over multiple GPUs with nearest-neighbor communications, often referred to as halo exchanges.

In the Livermore Big Artificial Neural Network (LBANN) deep learning framework, spatial-parallel convolution is implemented using several communication methods, including MPI and NVSHMEM. The MPI-based halo exchange uses the standard send and receive primitives, whereas the NVSHMEM-based implementation uses one-sided put, yielding significant performance improvements on Lawrence Livermore National Laboratory’s Sierra supercomputer.

Efficient Strong-Scaling on Sierra Supercomputer

Efficient Strong-Scaling on NVIDIA DGX SuperPOD

Accelerate Time to Solution

Reducing the time to solution for high-performance, scientific computing workloads generally requires a strong-scalable application. QUDA is a library for lattice quantum chromodynamics (QCD) on GPUs, and it’s used by the popular MIMD Lattice Computation (MILC) and Chroma codes.

NVSHMEM-enabled QUDA avoids CPU-GPU synchronization for communication, thereby reducing critical-path latencies and significantly improving strong-scaling efficiency.

Watch the GTC 2020 Talk

Simplify Development

The conjugate gradient (CG) method is a popular numerical approach to solving systems of linear equations, and CGSolve is an implementation of this method in the Kokkos programming model. The CGSolve kernel showcases the use of NVSHMEM as a building block for higher-level programming models like Kokkos.

NVSHMEM enables efficient multi-node and multi-GPU execution using Kokkos global array data structures without requiring explicit code for communication between GPUs. As a result, NVSHMEM-enabled Kokkos significantly simplifies development compared to using MPI and CUDA.

Productive Programming of Kokkos CGSolve