Ewa Deelman

Ewa Deelman is a Research Professor in the Department of Computer Science at the University of Southern California (USC) and a Principal Scientist at the USC Information Sciences Institute. She also directs Science Automation Technologies, a research group focused on automating scientific processes through advanced workflow management solutions. Her work is foundational in enabling complex scientific computations across distributed computing environments.

Deelman's research centers on scientific workflow automation, including resource provisioning, data management, and job scheduling in distributed systems. She pioneered workflow planning for distributed computations and leads the development of the Pegasus Workflow Management System. Current interests extend to anomaly detection using machine learning, quantum-classical hybrid workflows, and edge-to-cloud continuum integration. Her group's work bridges theoretical advances with practical tools for domain sciences like seismology and molecular dynamics.

Analysis of her recent publications reveals a strong emphasis on enhancing workflow robustness through machine learning and emerging technologies. Key trends include LLM-based scheduling optimization, pandemic-driven resilience studies, and quantum workflow integration. Her work increasingly addresses cross-cutting challenges in data-intensive science, particularly in managing computational workflows across heterogeneous infrastructures from edge devices to cloud platforms.

As Director of Science Automation Technologies, Deelman oversees a team that has made Pegasus a cornerstone tool in scientific computing. The system handles complex workflow orchestration for projects like CyberShake (seismic hazard analysis) and molecular dynamics simulations, automating data movement, execution, and error recovery across campus clusters, supercomputers, and cloud resources.