DASHMM: Dynamic Adaptive System for Hierarchical Multipole Methods

被引：12
|
作者：
DeBuhr, J. [1 ]
Zhang, B. [1 ]
Tsueda, A. [2 ]
Tilstra-Smith, V. [3 ]
Sterling, T. [1 ]
机构：
[1] Indiana Univ, Sch Informat & Comp, Ctr Res Extreme Scale Technol, Bloomington, IN 47404 USA
[2] Loyola Univ, Coll Arts & Sci, Chicago, IL 60660 USA
[3] Cent Coll, Dept Math & Phys, Pella, IA 50219 USA
来源：
COMMUNICATIONS IN COMPUTATIONAL PHYSICS | 2016年 / 20卷 / 04期
基金：
美国国家科学基金会;
关键词：
Barnes-Hut method; fast multipole method; Laplace potential; ParalleX; runtime software; PARALLEL IMPLEMENTATION; ALGORITHMS;
D O I：
10.4208/cicp.030316.310716sw
中图分类号：
O4 [物理学];
学科分类号：
0702 ;
摘要：
We present DASHMM, a general library implementing multipole methods (including both Barnes-Hut and the Fast Multipole Method). DASHMM relies on dynamic adaptive runtime techniques provided by the HPX-5 system to parallelize the resulting multipole moment computation. The result is a library that is easy-to-use, extensible, scalable, efficient, and portable. We present both the abstractions defined by DASHMM as well as the specific features of HPX-5 that allow the library to execute scalably and efficiently.
引用
收藏
页码：1106 / 1126
页数：21
相关论文
共 50 条
  • [1] Revision of DASHMM: Dynamic Adaptive System for Hierarchical Multipole Methods
    DeBuhr, J.
    Zhang, B.
    Sterling, T.
    COMMUNICATIONS IN COMPUTATIONAL PHYSICS, 2018, 23 (01) : 296 - 314
  • [2] DASHMM Accelerated Adaptive Fast Multipole Poisson-Boltzmann Solver on Distributed Memory Architecture
    Zhang, Bo
    DeBuhr, Jackson
    Niedzielski, Drake
    Mayolo, Silvio
    Lu, Benzhuo
    Sterling, Thomas
    COMMUNICATIONS IN COMPUTATIONAL PHYSICS, 2019, 25 (04) : 1235 - 1258
  • [3] Adaptive fast multipole methods on the GPU
    Anders Goude
    Stefan Engblom
    The Journal of Supercomputing, 2013, 63 : 897 - 918
  • [4] Adaptive fast multipole methods on the GPU
    Goude, Anders
    Engblom, Stefan
    JOURNAL OF SUPERCOMPUTING, 2013, 63 (03): : 897 - 918
  • [5] DYNAMIC AUTOTUNING OF ADAPTIVE FAST MULTIPOLE METHODS ON HYBRID MULTICORE CPU AND GPU SYSTEMS
    Holm, Marcus
    Engblom, Stefan
    Goude, Anders
    Holmgren, Sverker
    SIAM JOURNAL ON SCIENTIFIC COMPUTING, 2014, 36 (04): : C376 - C399
  • [6] Scalable Hierarchical Multipole Methods using an Asynchronous Many-Tasking Runtime System
    DeBuhr, Jackson
    Zhang, Bo
    D'Alessandro, Luke
    2017 IEEE INTERNATIONAL PARALLEL AND DISTRIBUTED PROCESSING SYMPOSIUM WORKSHOPS (IPDPSW), 2017, : 1226 - 1234
  • [7] Hierarchical Bayesian methods for estimation of parameters in a longitudinal HIV dynamic system
    Huang, Yangxin
    Liu, Dacheng
    Wu, Hulin
    BIOMETRICS, 2006, 62 (02) : 413 - 423
  • [8] Hierarchical adaptive dynamic power management
    Ren, ZY
    Krogh, BH
    Marculescu, R
    DESIGN, AUTOMATION AND TEST IN EUROPE CONFERENCE AND EXHIBITION, VOLS 1 AND 2, PROCEEDINGS, 2004, : 136 - 141
  • [9] Hierarchical adaptive dynamic power management
    Ren, ZY
    Krogh, BH
    Marculescu, R
    IEEE TRANSACTIONS ON COMPUTERS, 2005, 54 (04) : 409 - 420
  • [10] LOAD BALANCING AND DATA LOCALITY IN ADAPTIVE HIERARCHICAL N-BODY METHODS - BARNES-HUT, FAST MULTIPOLE, AND RADIOSITY
    SINGH, JP
    HOLT, C
    TOTSUKA, T
    GUPTA, A
    HENNESSY, J
    JOURNAL OF PARALLEL AND DISTRIBUTED COMPUTING, 1995, 27 (02) : 118 - 141
12345