Kaushik Velusamy

Kaushik Velusamy is an Assistant Computer Scientist in the Data Science Group at the Argonne Leadership Computing Facility, Argonne National Laboratory. His work bridges high-performance computing and machine learning, focusing on data system optimization for scientific applications.

Research Interests: His research spans optimizing data access performance for machine learning, scientific data management, and exploring novel computer architectures. Key technical areas include parallel I/O, distributed memory systems, collective communications, DAOS, HDF5, and large-scale distributed deep learning. He aims to improve efficiency and scalability in data-intensive scientific computing environments.

Education:

• Ph.D. in Computer Science, University of Maryland, 2021
• Advisors: Dr. Milton Halem, Dr. John Dorband

Professional Focus: His current projects include DLIO (Deep Learning I/O), performance characterization of deep learning workloads, and integration of high-performance storage systems with AI frameworks. His work supports scientific computing at leadership-class facilities.

Scientific Awards:

• No awards listed in the provided text.

Advising and Grants: No formal students or grant information is mentioned. He was advised during his Ph.D. by Dr. Milton Halem and Dr. John Dorband. No current advising or funding details are available.

Labs and Teams: He is a member of the Data Science Group at the Argonne Leadership Computing Facility (ALCF), a U.S. Department of Energy Office of Science user facility supporting cutting-edge computational science.