May 2021 OLCF User Conference Call: GPU Concurrency
The OLCF hosts monthly User Conference Calls. These calls are your opportunity to speak with center personnel to get the latest updates, express any concerns you may have, etc. No registration is required for this event.
Monthly Topic: GPU Concurrency
Speaker: Robert Searles, NVIDIA
Abstract: Concurrency refers to the ability to perform multiple CUDA operations simultaneously on the GPU. This allows the programmer to do things like run multiple kernels on a single GPU or spread their work across multiple GPUs. This presentation will present an overview of the different strategies that can be employed to achieve concurrency on NVIDIA GPUs including multi-threading, MPI, CUDA Multi-Process Service (MPS), and Multi-Instance GPU (MIG). There will also be additional CUDA Training Series modules offered as follow-ups to this presentation for participants who would like a deeper dive into any of the aforementioned topics.