Kevin A. Brown
Argonne National Laboratory (ANL)
Mathematics and Computer Science Division
Biography
Kevin A. Brown is the Assistant Computer Scientist at Argonne National Laboratory where he investigates new designs for supercomputer networks. He received his B.Sc. from the University of Technology, Jamaica (UTech) and his M.Sc. and Ph.D. in Mathematical and Computing Science from the Tokyo Institute of Technology. His prior work experience includes serving as a Unix Systems Administrator at Digicel (Jamaica) Ltd. and as a postdoctoral appointee in the Argonne Leadership Computing Facility focused on exascale interconnect performance evaluation.
SRP Project Title
Analyzing and Modeling Large Scale Infrastructure
Topical Areas
Artificial Intelligence and Intelligent Systems; Computer Science; Electrical, Electronic, and Information Engineering; High Performance Computing; Informatics, Analytics and Information Science; Infrastructure and Instrumentation; Other Computer and Information Sciences; Performance Evaluation and Benchmarking; Software Engineering; Statistics and Probability
Abstract
Large-scale computing infrastructureâincluding supercomputers and wide-area research networksâunderpins breakthroughs in science and engineering. Yet the scale and complexity of these systems make them difficult to deploy, operate, and optimize. They produce massive volumes of telemetry and logs that must be distilled into actionable insight, and faults can cascade across subsystems, degrading reliability and productivity if not detected and mitigated quickly. Our work brings together complementary capabilities to study, design, and optimize these infrastructures: ⢠System and workload modeling that combines AI, parallel discrete-event simulation (PDES), and fast surrogate models ⢠Scalable log and telemetry analysis, including automated, AI-assisted pipelines ⢠Anomaly detection and failure prediction to improve resilience ⢠Network performance benchmarking, measurement, and analysis to guide planning and operations
Desired Skills
- Basic programming experience is an asset
Lightning Talk Title
Analyzing and Modeling Large-Scale Infrastructure
Keywords
networking supercomputing HPC ESnet AI Artificial intelligence Resilience Failure analysis Performance analysis