Jeremy Kepner, Lincoln Laboratory Fellow in the Division of Cybersecurity and Information Sciences and Research Affiliate in the Mathematics Department at MIT, has been named in the 2021 class of Society for Industrial and Applied Mathematics (SIAM ). The Fellow designation honors SIAM members who have made outstanding contributions to the 17 fields of mathematics-related research that SIAM promotes through its publications, conferences and community of scientists. Kepner was recognized for “his contributions to interactive parallel computing, matrix graph algorithms, green supercomputing and big data”.
Since arriving at Lincoln Laboratory in 1998, Kepner has worked to expand computational capabilities in the laboratory and throughout the IT community. He has published extensively, served on technical committees of national conferences, and contributed to regional efforts to provide access to high-performance computing.
“Jeremy has contributed for twenty years to the important field of high performance computing, including supercomputers and embedded systems. It has also had a major impact on research on supercomputer systems. He invented a unique way of processing the signal on rare data, critically important for analyzing social networks and leading to more efficient use of parallel computing environments, ”says David Martinez, now Lincoln Laboratory fellow and previously chief of division that hired and then worked with Kepner for many years.
At Lincoln Laboratory, Kepner originally led the United States Department of Defense (DoD) high performance integrated computing software initiative that created the library standard for vector, signal, and image processing that many sensor systems of the DoD have used. In 1999 he invented MatlabMPI software and in 2001 was the architect of pMatlab (Parallel Matlab Toolbox) which has been used by thousands of Lincoln Laboratory staff, scientists and engineers around the world. In 2011, the Parallel Vector Tile Optimization Library (PVTOL), developed under the leadership of Kepner, won an R&D 100 award.
“Jeremy has been a global leader in the evolution of the state of high performance computing over the past two decades,” said Stephen Rejto, head of the cybersecurity and information science division at Lincoln Laboratory . “His vision and drive have been invaluable to the lab’s mission.”
Kepner led a consortium to pioneer the Massachusetts Green High Performance Computing Center, the largest in the world and, due to its use of hydropower, the most “green” open research data center, which enables dramatic increase in MIT’s computing capacity while reducing its CO2 footprint. He spearheaded the creation of what is now Lincoln Laboratory Supercomputing Center, which has New England’s most powerful supercomputer. In 2019, he helped create the U.S. Air Force-MIT Artificial Intelligence Accelerator, which leverages MIT and Air Force expertise and resources to advance artificial intelligence research.
“These individual accolades are recognition of the accomplishments of our entire Lincoln team to whom I am eternally indebted,” says Kepner.
Kepner’s recent work has focused on graphical analysis and Big Data. He created a new database management language and schema (Dynamic Distributed Dimensional Data Model, or D4M), which is widely used in Lincoln Laboratory and government big data systems.
His publications cover many areas: data mining, databases, high performance computing, graphics algorithms, cybersecurity, visualization, cloud computing, random matrix theory, abstract algebra and bioinformatics. Among his works are two bestselling SIAM books, “Parallel MATLAB for Multicore and Multinode Computers” and “Graph Algorithms in the Language of Linear Algebra”. In 2018, he and his co-author Hayden Jananthan published “Mathematics of Big Data” as one of the books in the MIT Lincoln Laboratory series published by MIT Press.
Kepner, who joined SIAM during his graduate studies at Princeton University, has not only published books and articles through SIAM, but has also been involved in the activities of the SIAM community. He was vice-president of the SIAM International Conference on Data Mining; advises a student section of SIAM; and enlisted SIAM’s affiliation to the High Performance Extreme (originally Embedded) Computing (HPEC) conference, in which it had “a pivotal role in bringing the high performance embedded computing community together and , under his leadership, became an IEEE conference in 2012, ”according to Martinez, who founded the HPEC conference hosted by the Lincoln Laboratory in 1997.
Kepner is the first Lincoln Laboratory researcher to achieve the rank of SIAM Fellow and the ninth from MIT.