Many GPU-accelerated HPC applications spend a substantial portion of their time in non-uniform, GPU-to-GPU communications, resulting in an increased solution times.
To ensure that GPU-to-GPU communication is as efficient as possible for HPC applications with non-uniform communication, it is crucial that these applications make informed decisions when assigning MPI processes to GPUs, ensuring processes requiring heavy communication use faster communication links.
NVIDIA Topology Aware GPU Selection (NVTAGS) is a toolset for HPC applications using MPI. It enables faster solve times for those applications with high GPU-communication-to-application runtime ratios by automatically profiling the application communication, extracting the system GPU communication topology leveraging NVIDIA System Management Interface (nvdia-smi), and finding an efficient process-to-GPU assignment that minimizes communication congestion.
Key benefits of NVTAGS:
- Automated process: NVTAGS provides a fully automated process that removes the need to conduct manual and time-consuming mapping of GPUs to processes.
- Improved performance: NVTAGS dramatically improves performance by intelligently mapping MPI processes to GPUs for HPC applications that require heavy GPU-to-GPU communication.
- Lightweight: It is extremely lightweight, with application profiling taking up less than 1% of the total application runtime. It is also very portable and can be easily deployed on a system or integrated within a container.