University of California, San Diego
Composable Systems for AI-Integrated Scientific Computing
Dr. İlkay Altıntaş is the Chief Data Science Officer of the San Diego Supercomputer Center as well as a Fellow of the Halıcıoğlu Data Science Institute at the University of California San Diego. She is the Founding Director of the Workflows for Data Science (WorDS) Center of Excellence and the WIFIRE Lab. The WoRDS Center specializes in the development of methods, cyberinfrastructure, and workflows for computational data science and its translation to practical applications. The WIFIRE Lab focuses on methods for all-hazards knowledge from data collection to modeling efforts, and has achieved significant success in helping to manage wildfires. Among the awards she has received are the 2015 IEEE TCSC Award for Excellence in Scalable Computing for Early Career Researchers and the 2017 ACM SIGHPC Emerging Woman Leader in Technical Computing Award. Altıntaş holds a Ph.D. degree from the University of Amsterdam in the Netherlands.
Scientific computing increasingly involves machine learning and artificial intelligence driven methods which require specialized capabilities for distributed data, networking and computing at the digital continuum. Such distributed architectures built around the composability of data-centric applications led to the emergence of a new ecosystem for container coordination and integration across dynamic resource pools. New approaches for dynamic composability of these emerging systems with traditional supercomputing systems are needed to further advance the data-driven scientific applications. This talk will present our approach for using composable systems in the intersection between scientific computing, artificial intelligence and sensor data integration. The architecture of a working example of a composable infrastructure that federates Expanse, an NSF-funded supercomputer, with Nautilus, a Kubernetes-based GPU geo-distributed cluster, and Sage, a cyberinfrastructure to enable AI at the edge, will be discussed with application case studies in hazards and Internet of Things. Case studies will be presented as integrated scientific workflows that bridge the insights from collaborative dynamic data-driven team science with heterogenous composable infrastructure.