Disruptive changes in high-performance computing hardware, where data motion is the dominant cost, present us with an opportunity to reshape how numbers are stored and moved in computing. We are developing methods, tools, and expertise to facilitate the use of mixed and dynamically adaptive precision numbers, allowing us to store and move the minimum number of bits required to perform a given calculation and help ease the data motion bottleneck on modern computer architectures.
Fox, A., G. Sanders, and A. Knyazev. 2018. "Investigation of Spectral Clustering for Signed Graph Matrix Representations" (forthcoming). IEEE HPEC, Waltham, MA, Sept. 2019. LLNL-CONF-754816.
Hoang, D. T., et al. 2018. "A Study of the Trade-off Between Reducing Precision and Reducing Resolution for Data Analysis and Visualization" (forthcoming). IEEE Visualization Vancouver, BC, Oct. 2019. LLNL-CONF-755005.
Lindstrom, P., S. Lloyd, and J. Hittinger. 2018. "Universal Coding of the Reals: Alternatives to IEEE Floating Point." Conference for Next Generation Arithmetic, Singapore, March 2018. LLNL-CONF-743265.
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.