Coursera Spring Sale
40% Off Coursera Plus Annual!
Grab it
This seminar talk by Sunita Chandrasekaran, Associate Professor at the University of Delaware, explores the integration of HPC and AI for optimizing plasma simulations. Learn about innovative strategies for streaming simulation data directly into machine learning frameworks, bypassing traditional file system bottlenecks and enabling in-transit data transformation. Discover how the PIConGPU (Plasma-in-Cell on GPU) application serves as a case study demonstrating the synergy between hardware innovation and software adaptation at scale. The presentation addresses the challenges of migrating legacy applications to new architectures and how the orchestration of HPC and AI workflows can accelerate scientific progress while improving computational efficiency. Chandrasekaran, who leads the Computational Research Programming Lab and co-directs the AI Center of Excellence, shares insights from her extensive experience in high performance computing, exascale computing, machine learning, and interdisciplinary science.
