With multidisciplinary expertise spanning technical pillars of high-performance computing, data science, and computational mathematics, we work toward building computational capabilities that position PNNL as a computing powerhouse. We also focus on enhancing the Science of Computing to achieve high-performance, power-efficient, and reliable computing at extreme scales for a spectrum of scientific endeavors that address significant problems of national interest, especially among PNNL’s core pursuitsenergy, the environment, national security, and fundamental science.
In the first study of its kind, scientists collaborating from the University of Washington, Warsaw University of Technology, Los Alamos National Laboratory and PNNL developed a novel model seeking a more intricate look at what happens during the final stages of the nuclear fission process. Using the model, they determined that fission fragments remain connected far longer than expected before the daughter nuclei split apart, delivering a long-awaited description of real-time fission dynamics within a microscopic framework and opening a pathway to a theoretical method with abundant predictive power. Notably, in addition to its publication, the paper was highlighted as an Editors’ Suggestion by Physical Review Letters.
Working with an international team of academics, Sriram Krishnamoorthy, a research scientist and System Software and Applications Team Lead in ACMD Division’s High Performance Computing group, will join a select group of scientists whose work will be featured as part of the 37th Annual ACM SIGPLAN Conference on Programming Language Design and Implementation, known as PLDI. This marks the fourth time in a row Krishnamoorthy’s work, this time focused on avoiding cache conflict misses that lower computing system performance, will be featured in this leading programming language research forum.
Shuaiwen Leon Song, with ACMD Division’s High Performance Computing group, co-authored two accepted full papers to be featured during this year’s ACM 25th International Symposium on High-Performance Parallel and Distributed Computing, known as HPDC’16. One paper details using novel algorithm-based fault tolerance schemes to detect and correct soft errors that can lead to silent data corruption, while the other aims to improve computing performance using a first-of-its-kind scheme that introduces false sharing in multithreaded systems to reduce memory contention and optimize HPC applications. Both papers will be featured at HPDC’16 being held May 31-June 04, 2016 in Kyoto, Japan.
Alexandre Tartakovsky, ACMD Division’s Associate Division Director for Computational Mathematics, was among a specially selected group of scientists who participated in the 20th German-American Kavli Frontiers of Science Symposium, held March 10-13, 2016 in Potsdam, Germany. The annual symposium is co-sponsored by the Alexander von Humboldt Foundation and National Academy of Sciences. Attendance is based on a selective process from among researchers who have made “recognized contributions to science, including recipients of major national fellowships and awards” or have been deemed “future leaders in science.” Notably, this marks the second consecutive year that Tartakovsky was invited to participate in the international symposium.
High-performance computing is an elemental part of bioinformatics, where software helps manage massive volumes of biological and genetic information. In “Parasail: SIMD C Library for Global, Semi-Global, and Local Pairwise Sequence Alignments,” author Jeff Daily, a scientist with ACMD Division’s HPC group, introduces the Parasail software library. Parasail represents the first time global, semi-global, and local alignment algorithms have been made available in one open-source high-performance software library, providing researchers with an option that improves upon earlier programs widely used in bioinformatics applications. The article about Parasail is featured in the February 2016 issue of BMC Bioinformatics.