PhD Opening: If you are looking for a PhD position and interested in this PhD subject, then please let me know.



About Me

I am an Inria (ISFP) researcher in the ROMA team at LIP, ENS Lyon. Previously, I was a postdoctoral researcher in the ALPINES team at Inria Paris. I was also a postdoctoral researcher at Pacific Northwest National Laboratory. I completed my PhD in 2017 at Inria Bordeaux under the supervision of Olivier Beaumont, Emmanuel Agullo, Lionel Eyraud-Dubois, and Samuel Thibault. After completing my Master's from IISc, Bangalore, and prior to joining PhD program, I worked at IBM India Research Lab for 17 months in High Performance Analytics group.

My main area of research is High Performance Computing (HPC). Presently I am interested in the design of parallel and scalable algorithms for tensor computations. A tensor is a generalization of vectors and matrices, and also can be viewed as a multidimensional array. It is common in HPC community to work with the matricized version of the tensor while performing different operations. The matricized version may hinder some of the multidimensional properties of the tensor. It may also increase data movement costs significantly when operating with different matricized versions of the same tensor. My approach is to consider tensors as multidimensional objects and extend existing (mostly matrix) techniques and algorithms for tensors. I also perform theoretical analysis (in terms of computations and communications) and implementations of the proposed algorithms for the state-of-the-art HPC systems. In past, I have worked on scheduling, design of new runtime systems, and performance analysis & optimization for heterogeneous systems. Most of my work was related to prove bounds for the scheduling algorithms or to obtain the maximum performance from the systems.

Teaching


Representative Publications

  1. Olivier Beaumont, Lionel Eyraud-Dubois, Suraj Kumar. Approximation Proofs of a Fast and Efficient List Scheduling Algorithm for Task-Based Runtime Systems on Multicores and GPUs [pdf]
    2. IEEE International Parallel & Distributed Processing Symposium (IPDPS 2017), May 2017, Orlando, Florida, USA.
  2. Olivier Beaumont, Terry Cojean, Lionel Eyraud-Dubois, Abdou Guermouche, Suraj Kumar. Scheduling of Linear Algebra Kernels on Multiple Heterogeneous Resources [pdf]
    3. International Conference on High Performance Computing, Data, and Analytics (HiPC 2016), Dec 2016, Hyderabad, India.
  3. Suraj Kumar, Lionel Eyraud-Dubois, Sriram Krishnamoorthy. Performance Models for Data Transfers: A Case Study with Molecular Chemistry Kernels [pdf]
    4. International Conference on Parallel Processing (ICPP 2019), August 2019, Kyoto, Japan.

Main Publications

  1. Hussam Al Daas, Grey Ballard, Laura Grigori, Suraj Kumar, Kathryn Rouse. Brief Announcement: Tight Memory-Independent Parallel Matrix Multiplication Communication Lower Bounds
    2. ACM Symposium on Parallelism in Algorithms and Architectures (SPAA 2022), Jul 2022, Philadelphia, PA, USA.
  2. Lionel Eyraud-Dubois, Suraj Kumar. Analysis of a List Scheduling Algorithm for Task Graphs on Two Types of Resources
    3. IEEE International Parallel & Distributed Processing Symposium (IPDPS 2020), May 2020, New Orleans (Virtual), Louisiana, USA.
  3. Suraj Kumar, Lionel Eyraud-Dubois, Sriram Krishnamoorthy. Performance Models for Data Transfers: A Case Study with Molecular Chemistry Kernels
    4. International Conference on Parallel Processing (ICPP 2019), August 2019, Kyoto, Japan.
  4. Olivier Beaumont, Lionel Eyraud-Dubois, Suraj Kumar. Fast Approximation Algorithms for Task-Based Runtime Systems
    5. Concurrency and Computation: Practice and Experience, Wiley, 2018, 30(17).
  5. Olivier Beaumont, Lionel Eyraud-Dubois, Suraj Kumar. Approximation Proofs of a Fast and Efficient List Scheduling Algorithm for Task-Based Runtime Systems on Multicores and GPUs
    6. IEEE International Parallel & Distributed Processing Symposium (IPDPS 2017), May 2017, Orlando, Florida, USA.
  6. Olivier Beaumont, Terry Cojean, Lionel Eyraud-Dubois, Abdou Guermouche, Suraj Kumar. Scheduling of Linear Algebra Kernels on Multiple Heterogeneous Resources
    7. International Conference on High Performance Computing, Data, and Analytics (HiPC 2016), Dec 2016, Hyderabad, India.
  7. Emmanuel Agullo, Olivier Beaumont, Lionel Eyraud-Dubois, Suraj Kumar. Are Static Schedules so Bad ? A Case Study on Cholesky Factorization
    8. IEEE International Parallel & Distributed Processing Symposium (IPDPS 2016), May 2016, Chicago, IL, United States. IEEE, 2016.
  8. Emmanuel Agullo, Olivier Beaumont, Lionel Eyraud-Dubois, Julien Herrmann, Suraj Kumar, Loris Marchal, Samuel Thibault. Bridging the Gap between Performance and Bounds of Cholesky Factorization on Heterogeneous Platforms
    9. Heterogeneity in Computing Workshop 2015, May 2015, Hyderabad, India. 2015.
  9. Ankur Narang, Suraj Kumar, Michael Perrone, David Wade, Kristian Bendiksen, Vidar Slatten, Tor Erik Rabben. Maximizing TTI RTM Throughput for CPU+ GPU
    10. 75th EAGE Conference & Exhibition incorporating SPE EUROPEC 2013.
  10. Ankur Narang, Suraj Kumar, Ananda S Das, Michael Perrone, David Wade, Kristian Bendiksen, Vidar Slatten, Tor Erik Rabben. Performance Optimizations for TTI RTM on GPU based Hybrid Architectures
    11. Biennial International Conference & Exposition, 2013.
  11. Anandhavalli Muniasamy, Suraj Kumar, Sudhanshu, Ayush Kumar, Mrinal K Ghose. Optimized association rule mining using genetic algorithm
    12. Advances in information mining, ISSN 9753265, Volume 1, Issue 2, 2009.

Publications (my contribution is limited)

  1. Karol Kowalski, Raymond Bair, Nicholas Bauman, Jeffery Boschen, Eric Bylaska, Jeff Daily, Wibe de Jong, Thom Dunning, Niranjan Govind, Robert Harrison, Marat Keceli, Kristopher Keipert, Sriram Krishnamoorthy, Suraj Kumar, Erdal Mutlu, Bruce Palmer, Ajay Panyala, Bo Peng, Ryan Richard, Tjerk Straatsma, Peter Sushko, Edward Valeev, Marat Valiev, Hubertus Van Dam, Jonathan Waldrop, David Williams-Young, Chao Yang, Marcin Zalewski, Theresa Windus. From NWChem to NWChemEx: Evolving with the Computational Chemistry Landscape
    2. Chemical Reviews 2021, Volume 121(8), 4962-4998.