logo
Università della Svizzera italiana
  • Italiano
  • English
  • usi.ch
  • Info Desk
  • Campus map
search.usi
Search for contacts, projects,
courses and publications
Italiano
People
    Education
    • Education
    • Courses
    Research
    • Projects
    • Publications
    • Competences maps
    Organisation
    • Faculties
    • Other organizational units

    Rethinking large-scale economic modeling for efficiency: optimizations for GPU and Xeon Phi clusters

    Additional information

    Authors
    Scheidegger S., Mikushin . D. ., Kuebler F., Schenk O.
    Type
    Article in conference proceedings
    Year
    2018
    Language
    English
    Abstract
    We propose a massively parallelized and optimized framework to solve high-dimensional dynamic stochastic economic models on modern GPU-and KNL-based clusters. First, we introduce a novel approach for adaptive sparse grid index compression alongside a surplus matrix reordering, which significantly reduces the global memory throughput of the compute kernels and maps randomly accessed data onto cache or fast shared memory. Second, we fully vectorize the compute kernels for AVX, AVX2 and AVX512 CPUs, respectively. Third, we develop a hybrid cluster oriented work-preempting scheduler based on TBB, which evenly distributes the time iteration workload onto available CPU cores and accelerators. Numerical experiments on Cray XC40 KNL "Grand Tave" and on Cray XC50 "Piz Daint" systems at the Swiss National Supercomputer Centre (CSCS) show that our framework scales nicely to at least 4,096 compute nodes, resulting in an overall speedup of more than four orders of magnitude compared to a single, optimized CPU thread. As an economic application, we compute global solutions to an annually calibrated stochastic public finance model with sixteen discrete, stochastic states with unprecedented performance.
    Conference proceedings
    IEEE International Parallel & Distributed Processing Symposium (IPDPS'18)
    Month
    May
    Publisher
    In Proceedings of the 32nd IEEE International Parallel & Distributed Processing Symposium (IPDPS'18)
    Start page number
    610
    End page number
    619
    Meeting name
    IEEE International Parallel & Distributed Processing Symposium (IPDPS'18)
    Meeting place
    Vancouver, Canada
    Meeting date
    May 25-29, 2018
    DOI
    10.1109/IPDPS.2018.00070
    Keywords
    High-Performance Computing, Macroeconomics, Public Finance, Adaptive Sparse Grids, Heterogeneous Systems, GPUs, MICs

    Faculties

    Faculty of Informatics

    Organizational units

    Institute of Computational Science (ICS)

    Università della
    Svizzera italiana

    Via Buffi 13
    6900 Lugano, Switzerland
    tel +41 58 666 40 00
    fax +41 58 666 46 47
    e-mail info@usi.ch
    Other contacts
    Feedback on the website

    Maps and directions

    • Lugano Campus
    • Mendrisio Campus
    • Bellinzona Campus

    Stay in touch

    • Facebook
    • Twitter
    • Instagram
    • Youtube
    • LinkedIn
    • Newsletter
    • Annual Report
    • Subscribe
    © Università della Svizzera italiana
    Disclaimer Credits
    swissuniversities.ch
    logo
    • Faculties
    • Institutes
    • Bodies
    • Libraries and archives
    • Areas
    • Services
    • Job offers