• Phone: (925) 423-8454
• E-mail: devarajan1@llnl.gov

Education

• BTech in Computer Science, National Institute of Technology, Jamshedpur India, 2013
• PhD in Computer Science, Illinois Institute of Technology, Chicago, 2021

Hariharan Devarajan is a computer scientist in the Parallel Systems Group in the Center for Applied Scientific Computing (CASC) at Lawrence Livermore National Laboratory. His research on large-scale distributed systems focuses on data management for extreme-scale systems. His other research interests include scalable performance analysis, bottleneck detection and tuning, and data management in distributed deep-learning workloads. Hariharan has worked at LLNL since 2021 and has received best paper awards at HPDC and CCGrid conferences.

He received his PhD in computer science in 2021 from the Illinois Institute of Technology, Chicago under the guidance of his advisor, Xian-He Sun. His academic journey began with a BTech in computer science and engineering in 2013 from the National Institute of Technology, Jamshedpur India. After his undergrad, Hariharan gained practical experience as an application developer for Oracle Banking Platform in Mumbai, India.

Hariharan's current research focuses on workload-aware storage for HPC systems. He contributes to several projects within the laboratory, such as UnifyFS, the Scalable Checkpoint/Restart Library (SCR) project, FRACTALE, and GOTCHA. He is the task lead for I/O in the FRACTALE project, guiding research directions for data management for MD workflows, describing complex I/O patterns within the Flux scheduler, and improving utilization through data-aware job scheduling. These projects work towards a fundamental understanding of current and emerging needs for data management in HPC systems.