We plan to develop a solution to the well-known data placement problem caused by the growing data processing demands of increasingly complex and heterogeneous computing systems. The resulting compiler and runtime techniques will ensure that the applications that support national missions can be adapted to emerging extreme-scale architectures with complex memory hierarchies.
Bari, M. et al. 2018. "Is Data Placement Optimization Still Relevant on Newer GPUs?" Performance Modeling, Benchmarking, and Simulation of High Performance Computer Systems, Dallas, TX, Nov. 2018. LLNL-CONF-757796.
Stoltzfus, L. et al. 2018. "Data Placement Optimization in GPU Memory Hierarchy Using Predictive Modeling." Workshop on Memory-Centric High Performance Computing. Dallas, TX, Nov. 2018. LLNL-CONF-758021.
Lawrence Livermore National Laboratory • 7000 East Avenue • Livermore, CA 94550
Operated by Lawrence Livermore National Security, LLC, for the Department of Energy's National Nuclear Security Administration.