Learn about InfiniBand, In-Network Computing, and BlueField DPU capabilities in this 25-minute technical talk from NVIDIA experts Gilad Shainer and Richard Graham. Explore how complex scientific workloads are creating unprecedented demands on supercomputing infrastructures and discover solutions for cloud providers who need to deliver bare-metal performance for scientific applications. Dive into the programmable networking platform that incorporates hardware acceleration engines to reduce time to insight and boost overall performance. Gain valuable insights into how these technologies work together to drive innovation in scientific computing and handle the increasing complexities of modern computational workloads.
Overview
Syllabus
Accelerating Scientific Computing Workloads within InfiniBand DPUs
Taught by
OpenFabrics Alliance
Related Courses
-
Introduction to NVIDIA DOCA - Accelerating Data Center Applications with BlueField DPU
-
Leveraging SmartNICs for HPC and Data Center Applications - Part 2
-
Leveraging SmartNICs for HPC and Data Center Applications - Part 1
-
Programming Novel AI Accelerators for Scientific Computing - Part 1 - Tutorial 5
-
Programming Novel AI Accelerators for Scientific Computing - Part 2 - Tutorial 5
-
Accelerating Apache Spark Workloads with Apache DataFusion Comet
