High-performance computing expert Kumar joins CS faculty
High-performance computing expert Kumar joins CS faculty Heading link
Sidharth Kumar joined the CS department this fall as an assistant professor. His research lies at the intersection of high-performance computing, data analytics, and artificial intelligence (AI).
Kumar earned his bachelor’s degree from Dhirubhai Ambani Institute of Information and Communication Technology in India. He went to the University of Utah intending to earn his master’s degree in computing. While at the University of Utah, Kumar visited Argonne National Laboratory, where he was first exposed to high-performance computing (HPC).
“I really loved working on supercomputers solving real-world problems, and that’s when I decided I should do a PhD,” Kumar said.
At Argonne, Kumar worked closely with Mike Papka, who earned his PhD at UIC and is now a professor in UIC’s computer science department. Kumar spent three years interning at Argonne and has continued a collaboration with the lab.
After his postdoc, Kumar spent five years as an assistant professor at the University of Alabama at Birmingham before joining UIC.
“I’m excited to be in a much bigger department at UIC, and the proximity to Argonne was important,” Kumar said. “There are more opportunities for collaborations here. There are a lot of people working in different areas of computer science.”
Kumar works on problems in parallel computing, big data processing, parallel inputs/outputs (I/O), scalable algorithms, large data file formats, and performance modeling.
“While everyone has been fascinated by what AI/ML has been able to achieve in the past decade, one often overlooks the systems aspects — the underrated behind-the-scenes guy,” Kumar said. “It takes an army of CPUs/GPUs to train large-scale machine-learning models. All major companies like Google, Microsoft, and Meta are investing in an AI supercomputer. With my research, I aim to develop core algorithmic techniques to optimize the performance of irregular applications such as ML training on large-scale supercomputers.”
Kumar is teaching CS 480, Database Systems, this fall. He enjoys teaching and looks forward to working with UIC students.
“I’ve realized that I’m more passionate about teaching than I thought I would be,” Kumar said. “I love interacting with students in general. My classes are very interactive as I love talking to students while I’m teaching.”
Kumar is in the process of bringing his three graduate students to UIC and is looking forward to finding additional graduate students here.
His research has been published at major HPC conferences, including SC, HPDC, IPDPS, Cluster, CCGrid, ICPP, and HiPC. His research in parallel I/O has been deployed at the highest scale (768K cores) and also put into production runs (at 260K cores) at some of the fastest supercomputers in the world (Theta, Mira, Polaris, Shaheen, Edison, and Hopper).
His recent work in parallelizing relational algebra primitives to support declarative programming garnered the best research paper award at HiPC and the Hans Meuer best research paper award at ISC and was among a few researchers highlighted by Argonne National Laboratory’s yearly science magazine. His current research is supported by four NSF grants and one NIH grant.