Summary of Research
Fault tolerance is a fundamental issue in High Performance Computing, and while the challenges are exacerbated when computations are distributed across multiple compute nodes, software-based solutions are achievable. The goal of my research is to design an automated approach for providing fault tolerance for user applications, perhaps via application of program synthesis, an application of machine learning that synthesizes programs from high level specifications, or regular compile-time transformation. In distributed computing, the Partitioned Global Address Space (PGAS) memory model allows for accesses of memory that is 'remote', or allocated on another compute node, via one-sided Remote Direct Memory Access (RDMA) operations known as PUT and GET. Chapel, Cray's exascale and PGAS language, created as part of DARPA's High Productivity Computing Systems program, is highly receptive to change and can serve as an ideal test bed for research and experimentation. As a bonus, any solutions designed for Chapel may be generalizable enough to be implemented in other PGAS languages, or to expose new paradigms for software-based fault tolerance in distributed computing that may work even under MPI.
Louis Jenkins, Tingzhe Zhou, and Michael Spear. "Redesigning Go's Built-In Map to Support Concurrent Operations." 26th International Conference on Parallel Architectures and Compilation Techniques (PACT), IEEE, 2017.
Louis Jenkins. "RCUArray: An RCU-Like Parallel-Safe Distributed Resizable Array." 2018 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW). IEEE, 2018.
G. Dewan and L. Jenkins, "Paving the way for Distributed Non-Blocking Algorithms and Data Structures in the Partitioned Global Address Space model," 2020 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW), New Orleans, LA, USA, 2020, pp. 659-666, doi: 10.1109/IPDPSW50202.2020.00111.
Louis Jenkins, Marcin Zalewski, and Michael Ferguson. "Chapel Aggregation Library (CAL)." To Appear @ "Supercomputing 2018: Parallel Applications Workshop - Alternatives to MPI"
Louis Jenkins, et al. "Chapel HyperGraph Library (CHGL)." 2018 IEEE High Performance Extreme Computing Conference (HPEC). IEEE, 2018.
Joslyn, Cliff A., Sinan Aksoy, Dustin Arendt, Jesun Firoz, Louis Jenkins, Brenda Praggastis, Emilie Purvine, and Marcin Zalewski. "Hypergraph analytics of domain name system relationships." In International Workshop on Algorithms and Models for the Web-Graph, pp. 1-15. Springer, Cham, 2020.
J. S. Firoz, L. Jenkins, C. Joslyn, B. Praggastis, E. Purvine and M. Raugas, "Computing Hypergraph Homology in Chapel," 2020 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW), New Orleans, LA, USA, 2020, pp. 667-670, doi: 10.1109/IPDPSW50202.2020.00112.
Louis Jenkins, et al. "Graph Algorithms in PGAS: Chapel and UPC++" 2019 IEEE High Performance Extreme Computing Conference (HPEC). IEEE, 2019.
Louis Jenkins, et al. "Chapel HyperGraph Library (CHGL)", presented at Richland, WA, URL: http://louisjenkinscs.github.io/posters/CHGL.pdf
Departmental Outstanding Graduate Teaching Assistant Award @ University of Rochester (2020)
Outstanding Performance Award @ Pacific Northwest National Laboratory (2019)
Peer's Choice for Outstanding Project @ Lehigh University (2016)
Honorable Mention @ Computing Research Association's Outstanding Undergraduate Researcher's Competition, 2017
Dean's List @ Bloomsburg University, Spring 2014, Fall 2015, Spring 2017