Title | ||
---|---|---|
Evaluating the networking characteristics of the Cray XC-40 Intel Knights Landing-based Cori supercomputer at NERSC. |
Abstract | ||
---|---|---|
There are many potential issues associated with deploying the Intel Xeon Phi(TM) (code named Knights Landing [KNL]) manycore processor in a large-scale supercomputer. One in particular is the ability to fully utilize the high-speed communications network, given that the serial performance of a Xeon Phi(TM) core is a fraction of a Xeon (R) core. In this paper, we take a look at the trade-offs associated with allocating enough cores to fully utilize the Aries high-speed network versus cores dedicated to computation, eg, the trade-off between MPI and OpenMP. In addition, we evaluate new features of Cray MPI in support of KNL, such as internode optimizations. We also evaluate one-sided programming models such as Unified Parallel C. We quantify the impact of the above trade-offs and features using a suite of National Energy Research Scientific Computing Center applications. |
Year | DOI | Venue |
---|---|---|
2018 | 10.1002/cpe.4297 | CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE |
Keywords | Field | DocType |
high-performance computing,high-speed interconnects,Knights Landing,MPI,performance analysis,UPC | Manycore processor,Programming paradigm,Suite,Supercomputer,Computer science,Parallel computing,Xeon,Operating system,Computation | Journal |
Volume | Issue | ISSN |
30 | SP1 | 1532-0626 |
Citations | PageRank | References |
4 | 0.45 | 4 |
Authors | ||
6 |
Name | Order | Citations | PageRank |
---|---|---|---|
Douglas Doerfler | 1 | 105 | 9.48 |
Brian Austin | 2 | 33 | 5.23 |
Brandon G. Cook | 3 | 15 | 4.95 |
Jack Deslippe | 4 | 78 | 12.45 |
Krishna Kandalla | 5 | 4 | 0.45 |
Peter Mendygral | 6 | 23 | 1.87 |