Paper Title: Optimizing Edge AI, Networking and Storage by Combining GPU and DPU Technology

Paper Abstract: Edge is the new frontier of compute and data in today's world, driven by the explosion of mobile devices, work from home, digital video, smart cities, and connected cars. An increasing percentage of data generation, use and analysis now happens at the edge of the network, as does an increasing share of AI computing. These changes require faster computing, storage, and networking at the edge, and between the edge and the data center, or from the edge to the cloud. GPUs have proven to be the winning solution for AI and DPUs as they are revolutionizing infrastructure processing including networking, storage, and security. A new approach is to combine GPU and DPU technologies onto one card, delivering faster and more efficient AI and visual processing at the Edge. This is ideal for 5G wireless, cybersecurity, and omniverse types of solutions.

Paper Author: John Kim, Director of Storage Marketing, NVIDIA

Author Bio: John Kim is Director of Storage Marketing at NVIDIA in the Networking Business Unit, where he helps customers and vendors benefit from high performance network connections, SmartNIC offloads, and Data Processing Units, especially in the areas of storage, big data, and artificial intelligence. A frequent blogger, conference speaker, and webcast presenter, John is also chair of the Storage Networking Industry Association’s Networking Storage Forum (SNIA NSF). After starting his high tech career at an IT helpdesk and as a network administrator, John worked in solution marketing, product management, and alliances for enterprise software companies and then at enterprise storage vendors NetApp and EMC. He joined Mellanox in 2013 and then NVIDIA in 2020.