We focus on merging high-performance computing with data-centric analysis capabilities to solve significant problems in energy, the environment, and national security. PNNL has made scientific breakthroughs and advanced frontiers in high-performance computer science, computational biology and bioinformatics, subsurface simulation modeling, and multiscale mathematics.
During GraphLab Conference 2014, Sutanay Choudhury, a research scientist with PNNL’s Data Sciences group (ACMD Division), hosted a demonstration showcasing M&Ms4Graphs, a graph analytics framework for cyber security. M&Ms4Graphs uses graph-theoretic models to provide continuous updates on system states as part of enabling a resilient (a system’s ability to function in the face of impediments) cyber infrastructure. The project is one of many backed by PNNL’s Asymmetric Resilient Cybersecurity Initiative and features a diverse team of computer scientists and mathematicians from both PNNL’s Fundamental & Computational Sciences and National Security directorates, including major contributors Peter Hui, Kiri Oler, Chase Dowling, Emilie Hogan, and Mahantesh Halappanavar.
This month, Dr. Alexandre Tartakovsky joins the Advanced Computing, Mathematics, and Data Division as the full-time Associate Division Director for Computational Mathematics. In his new role, Alex will oversee the talented personnel who compose ACMD Division’s Computational Mathematics group, which includes computational engineering, uncertainty quantification, multiscale mathematics, and computational social sciences teams. Alex’s goal is to continue building PNNL’s Computational Mathematics group to world-class strength. He also will continue his scientific leadership role in applied mathematics through a variety of important projects.
In a demonstration of the Laboratory's ongoing commitment to enhancing the Science of Computing by achieving high-performance, power-efficient, and reliable computing at extreme scales, scientists from PNNL's HPC group have joined researchers from the U.S. Army Research Communications-Electronics Research, Development, and Engineering Center (Night Vision and Electronic Sensors Directorate) and the Defense Advanced Research Projects Agency (Microsystems Technology Office) as part of the Suite of Embedded Applications and Kernels, or SEAK, program. SEAK's primary goal is to advance the capabilities of high-performance embedded computing applications for U.S. Department of Defense purposes. PNNL is handling all of the SEAK program's technical aspects, led by Adolfy Hoisie, the project's principal investigator.
Tensor contractions, generalized matrix multiplications that are time-consuming to calculate, make them among the most compute-intensive operations in several ab initio computational quantum chemistry methods. In this work, Sriram Krishnamoorthy, a research scientist with ACMD’s High Performance Computing group, along with scientists from The Ohio State University, developed a systematic framework that uses three fundamental communication operators—recursive broadcast, rotation, and reduction—to derive communication-efficient algorithms for distributed contraction of arbitrary dimensional tensors on the IBM Blue Gene/Q Mira supercomputer. The framework automatically models potential space-performance trade-offs to optimize the communication costs incurred in executing tensor contractions on supercomputers. The paper documenting this work, “Communication-optimal Framework for Contracting Distributed Tensors,” is a SC14 Best Paper award finalist.
Congratulations to Zhijie “Jay” Xu, of PNNL’s Multiscale Mathematics team (ACMD), who was named an editorial board member of the International Journal of Computational Mathematics, a peer-reviewed, open-access journal that publishes original research and review articles spanning all areas of computational mathematics. He joins a 25-person international board with diverse mathematics-based expertise and is the sole representative from a U.S. Department of Energy national laboratory.