posted on 2019-05-15, 00:24authored byBarbara Chapman
Today’s High Performance Computing architectures exhibit significant compute power within each
node of the machine, often achieved via the inclusion of one or more accelerators that are attached
to CPUs. As a result, it has become essential that large-scale applications make effective use of
intra-node as well as inter-node parallelism. In the U.S. Department of Energy’s Exascale
Computing Project, several different approaches are being developed to support this requirement.
Of these, the most widely adopted so far is OpenMP, a directive-based parallel programming
interface supported by many compilers for Fortran, C and C++. In this presentation we discuss the
challenges of intra-node programming and how OpenMP attempts to meet them.
ABOUT THE AUTHOR
Barbara Chapman is a Professor of Applied Mathematics and Statistics, and of Computer Science,
at Stony Brook University, where she is affiliated with the Institute for Advanced Computational
Science. She also directs Computer Science and Mathematics Research at Brookhaven National
Laboratory. Barbara performs research on parallel programming interfaces and the related
implementation technology, and has been involved in several efforts to develop community
standards for parallel programming, including OpenMP, OpenACC and OpenSHMEM. Her
research group has created an open source compiler, OpenUH, that enabled practical
experimentation with proposed enhancements to application programming interfaces and a
reference implementation of the library-based OpenSHMEM standard. Dr. Chapman has coauthored over 200 papers and two books. She obtained her B.Sc. Hons in Mathematics at the
University of Canterbury and her Ph.D. in Computer Science from Queen’s University of Belfast.