
Collects and packages NVSHMEM put/get perftest results with system and topology evidence
- Open source
- 4K GitHub stars
- Developer tools


Collects and packages NVSHMEM put/get perftest results with system and topology evidence

Guides an agent to find CUDA memory, shared-memory race and sync bugs with Compute

Plans and validates NVSHMEM and NVSHMEM4Py installs across packages, containers and source

Looks up version-aware official NVSHMEM and NVSHMEM4Py documentation for agents

Diagnoses why an NVFLARE federated learning job failed, stalled or lost clients

Recommends NVSHMEM NIC-to-PE mappings and environment exports from GPU and RDMA topology

Diagnoses and narrowly repairs one reproduced NVIDIA PVA operator failure

Chooses an NVSHMEM remote transport for a target system using read-only evidence

Guides an agent to fix one known local bug with a patch commit and proof report

Guides beginners through NVSHMEM concepts, first C/C++ or Python programs, and launching

Turns a vague or intermittent bug report into a minimal, verified reproduction

Turns rough bug notes, error logs and screenshots into structured GitHub issue reports
The closest alternatives to Nvshmem troubleshoot and report bugs are Nvshmem collect performance data, Compute sanitizer diagnose cuda bugs, Nvshmem install, Nvshmem docs and Nvflare diagnose job.
Yes. Nvshmem collect performance data, Compute sanitizer diagnose cuda bugs, Nvshmem install have a free or open-source option.