I am a computer scientist at the Center for Computing Research at Sandia National Laboratories. I have been in Sandia since 2010.
Ph.D. CISE, University of Florida, 2009. Advisor: Prof. Tim Davis
My primary focus is on High Performance Computing. Within this broad field I focus on
- Linear solvers: Sparse factorizations and preconditioners such as communication-avoiding preconditioners, hybrid (direct + iterative) preconditioners and direct solvers
- Combinatorial algorithms for scientific computing problems: Graph/Hypergraph algorithms for partitioning, sparse matrix ordering and matching problems
- Graph algorithms for analytics: Graph traversals, finding connected components, graph analysis and partitioning of small-world graphs
- Linear Algebra Kernels: Sparse matrix-matrix multiplication, Batched dense linear algebra kernels
- Applications : Integrating the above methods in applications of importance
For more details, visit my research page.
- Paper on FROSch: A Fast and Robust Overlapping Schwarz Domain Decomposition Preconditioner Based on Xpetra in Trilinos., got accepted to the proceedings of the Domain Decomposition Conference. (with Alexander Heinlein, Axel Klawonn and Oliver Rheinbach).
- Paper on Geometric Mapping of Tasks to Processors on Parallel Computers with Mesh or Torus Networks, got accepted to IEEE Transactions on Parallel and Distributed Systems. (with Mehmet Deveci, Karen Devine, Kevin Pedretti, Mark Taylor, and Umit V Catalyurek).
- Report on Asynchronous One-Level and Two-Level Domain Decomposition Solvers, got posted on arXiv. (with Christian Glusa, Paritosh Ramanan, Erik Boman, and Edmond Chow).
- Paper on Fast Triangle Counting Using Cilk, got accepted to IEEE HPEC Graph Challenge. (with Abdurrahman Yasar, Michael Wolf, Jonathan Berry and Umit Catalyurek).
- Selected as the Graph Challenge 2018 Champions.
- Paper on Employing Multiple Levels of Parallelism for CFD at Large Scales on Next Generation High-Performance Computing Platforms, appeared on the proceedings of ICCFD10. (with Micah Howard, Travis Fisher, Mark Hoemmen, D. Dinzl, James Overfelt, Andrew Bradley, and Kyungjoo Kim).
- Paper on Multithreaded Sparse Matrix-Matrix Multiplication for Many-Core and GPU Architectures, got accepted to Parallel computing. (with Mehmet Deveci and Christian Trott).
- Report on Geometric Partitioning and Ordering Strategies for Task Mapping on Parallel Computers, got posted on arXiv. (with Mehmet Deveci, Karen Devine, Kevin Pedretti, Mark taylor, and Umit Catalyurek).
- Report on Sparse Matrix-Matrix Multiplication on Multilevel Memory Architectures : Algorithms and Experiments, got posted on arXiv. (with Mehmet Deveci, Si Hammond, and Michael Wolf).
- Paper on Experimental Design of Work Chunking for Graph Algorithms on High Bandwidth Memory Architectures, got accepted to IPDPS 2018. (with George Slota).
- Paper on Tacho: Memory-Scalable Task Parallel Sparse Cholesky Factorization, got accepted to AsHES workshop at IPDPS 2018. (with Kyungjoo Kim and Carter Edwards).
- Older talks, announcements and past news are archived.
(505) 844-7181 (Phone)
Mailing address (USPS)
Sandia National Laboratories
P.O. Box 5800, MS 1320
Albuquerque, NM 87185-1320
Sandia National Laboratories
1515 Eubank SE
Albuquerque, NM 87123