Hari Sundar


Photo of hari

Hari Sundar

Assistant Professor
School of Computing
University of Utah

email:  hari at cs dot utah dot edu
office:  3454 MEB
phone: (801) 585 9913


Research

The central focus of my research is the development of computationally optimal parallel, high-performance algorithms, both discrete and continuous, that are efficient and scalable on state-of-the-art architectures. It is driven by applications in biosciences, computational relativity and geophysics, such as cardiovascular mechanics, binary black-hole mergers, and seismic wave propagation. My research has resulted in the development of state-of-the-art distributed algorithms for adaptive mesh refinement, geometric multigrid, fast Gauss transform and sorting.

Recent & Upcoming Talks

  1. Parallel Fast Gauss Transform, SIAM PP'18, Mar 7, Waseda University, Tokyo, Japan.
  2. Efficient Parallel Streaming Algorithms for large-scale Inverse Problems - September 13, 2017 - 2017 IEEE High Performance Extreme Computing Conference, Waltham, MA
  3. Parallel Algorithms for the Computation of Cycles in Relative Neighborhood Graphs - August 16, 2017 - 46th International Conference on Parallel Processing, Bristol, UK

Recent Papers

Here are some recent papers. Find a complete list here.

  1. Hari Sundar, Efficient Parallel Streaming Algorithms for large-scale Inverse Problems, 2017 IEEE High Performance Extreme Computing Conference (HPEC ‘17), 2017.
  2. Isuru Fernando, Sanath Jayasena, Milinda Fernando, Hari Sundar, A Scalable Hierarchical Semi-Separable Library for Heterogeneous Clusters, Parallel Processing (ICPP), 46th International Conference on, 513-522, 2017.
  3. Parmeshwar Khurd, Hari Sundar, Parallel Algorithm for the Computation of Cycles in Relative Neighborhood Graphs, Parallel Processing (ICPP), 46th International Conference on, 191-200, 2017.
  4. Milinda Fernando, Dmitry Duplyakin, Hari Sundar, Machine and Application Aware Partitioning for Adaptive Mesh Refinement Applications, Proceedings of the 26th International Symposium on High-Performance Parallel and Distributed Computing (HPDC'17), 2017.
  5. Amir Gholami, Dhairya Malhotra, Hari Sundar, George Biros, FFT, FMM, or Multigrid? A comparative study of state-of-the-art Poisson solvers in the unit cube, SIAM Journal on Scientific Computing 38(3), C280–C306. 2016.

Students

Current

  1. Maxx Carlson - MS
  2. Weerahannadige Milinda Shyamala Fernando - PhD
  3. Seyed Majid Rasouli-Pichahi - PhD
  4. Nishit Tirpankar - PhD

Alumni

  1. Bryant Baird - MS 2017 - Adobe
  2. Christopher Mertin - MS 2017 - IBM
  3. Vishal Sharma (with G. Gopalakrishnan) - PhD 2016 - Microsoft

Prospective Students

I am interested in taking on new students (undergrad, masters & PhD). If you are interested in working on the kind of problems that interest me, feel free to email me. Currently, I am particularly interested in students with experience in CUDA/GPU Programming.