Dhairya Malhotra, a 2017 CSEM alumnus who studied with ICES Professor George Biros, has won the inaugural Association for Computing Machinery (ACM), Special Interest Group on High Performance Computing's (SIGHPC) Doctoral Dissertation Award.
SIGHPC based winners on technical depth, significance of the research contribution, potential impact on theory and practice, and overall quality of work.
SIGHPC reports that Malhotra’s dissertation, entitled Fast integral equation solver for variable coefficient elliptic PDEs in complex geometries, presents a parallel software framework applicable to a wide range of problems in fluid mechanics and electrostatics, offering up to an order of magnitude speed-up over the state of the art for solving complex fluid problems. In his research, Malhotra worked on all aspects of developing high quality mathematical software: novel mathematical formulations, convergence analysis, algorithm design, performance analysis, single-core and node optimization, correctness, robustness, and software dissemination for truly challenging problems.
“We are excited to grant the first Dissertation Award to Dhairya. His work exemplifies the kind of HPC research that SIGHPC supports, and helps to raise the standards of the profession,” said Dr. Jeffrey Hollingsworth, SIGHPC Chair after the selection. “It’s encouraging that the submissions for this first award were of such a high caliber. It would be great to be inundated with quality submissions in the future.”
Malhotra is currently a postdoctoral associate at the Courant Institute of Mathematical Sciences at New York University. He is also a past recipient of the ACM-IEEE George Michael Memorial HPC Fellowship, honoring exceptional Ph.D. students throughout the world whose research focus areas are in high performance computing, networking, storage, and large-scale data analysis.
“I am grateful to SIGHPC for the recognition of my efforts,” said Malhotra to SIGHPC after learning of the award. “The technologies and the software tools that I have developed with my advisor Dr. Biros, like our parallel fast multipole method library, are being used by other researchers and will continue to be a great resource for many computational scientists.”