Paper Title: Accelerating the Data Path to the GPU for AI and Beyond

Paper Abstract: As workflows shift away from the CPU in GPU-centric systems, the data path from storage to GPUs increasingly becomes the bottleneck. NVIDIA and its partners are relieving that bottleneck with a technology called GPUDirect Storage that includes a new set of interfaces. GPUDirect Storage enables acceleration of AI/ML and other applications by removing storage I/O bottlenecks into or out of the GPU. It introduces a new set of APIs as part CUDA to accelerate GPU-based applications, and applications using higher level frameworks see benefits with only a few lines of code changes, if any. Further, APIs (C/C++/Python) are provided for more control if necessary. This talk will cover the technology behind the acceleration, frameworks, and benefits for the end user applications.

Paper Author: Sandeep Joshi, Sr Manager, Nvidia

Author Bio: Sr. Manager at Nvidia who has background and experience in storage and database industry. MS, Santa Clara University.