Automatic differentiation of C++ codes for large-scale scientific computing

被引：0
|
作者：
Bartlett, Roscoe A. [1 ]
Gay, David M. [1 ]
Phipps, Eric T. [1 ]
机构：
[1] Sandia Natl Labs, Albuquerque, NM 87185 USA
来源：
COMPUTATIONAL SCIENCE - ICCS 2006, PT 4, PROCEEDINGS | 2006年 / 3994卷
关键词：
D O I：
暂无
中图分类号：
TP301 [理论、方法];
学科分类号：
081202 ;
摘要：
We discuss computing first derivatives for models based on elements, such as large-scale finite-element PDE discretizations, implemented in the C++ programming language. We use a hybrid technique of automatic differentiation (AD) and manual assembly, with local element-level derivatives computed via AD and manually summed into the global derivative. C++ templating and operator overloading work well for both forward- and reverse-mode derivative computations. We found that AD derivative computations compared favorably in time to finite differencing for a scalable finite-element discretization of a convection-diffusion problem in two dimensions.
引用
收藏
页码：525 / 532
页数：8
相关论文
共 50 条
  • [41] The Large-scale Structure of Scientific Method
    Kosso, Peter
    [J]. SCIENCE & EDUCATION, 2009, 18 (01) : 33 - 42
  • [42] The Large-scale Structure of Scientific Method
    Peter Kosso
    [J]. Science & Education, 2009, 18 : 33 - 42
  • [43] Model predictive analysis for autonomic workflow management in large-scale scientific computing environments
    Nordstrom, Steve
    Dubey, Abhishek
    Keskinpala, Turker
    Datta, Rahul
    Neema, Sandeep
    Bapty, Ted
    [J]. FOURTH IEEE INTERNATIONAL WORKSHOP ON ENGINEERING OF AUTONOMIC & AUTONOMOUS SYSTEMS, PROCEEDINGS, 2007, : 37 - +
  • [44] A CLOUD COMPUTING PLATFORM FOR LARGE-SCALE FORENSIC COMPUTING
    Roussev, Vassil
    Wang, Liqiang
    Richard, Golden
    Marziale, Lodovico
    [J]. ADVANCES IN DIGITAL FORENSICS V, 2009, 306 : 201 - 214
  • [45] Editors' preface for the topical issue "Numerical Methods for Large-Scale Scientific Computing, II"
    Karatson, Janos
    Korotov, Sergey
    Margenov, Svetozar
    [J]. CENTRAL EUROPEAN JOURNAL OF MATHEMATICS, 2013, 11 (08): : 1359 - 1360
  • [46] Web portal to make large-scale scientific computations based on Grid computing and MPI
    Akzhalova, Assel Zh.
    Aizhulov, Daniar Y.
    [J]. PARALLEL PROCESSING AND APPLIED MATHEMATICS, 2008, 4967 : 888 - 893
  • [47] Component-based, problem-solving environments for large-scale scientific computing
    Johnson, C
    Parker, S
    Weinstein, D
    Heffernan, S
    [J]. CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE, 2002, 14 (13-15): : 1337 - 1349
  • [48] Editors' preface for the topical issue "Numerical Methods for Large-Scale Scientific Computing, I"
    Karatson, Janos
    Korotov, Sergey
    Margenov, Svetozar
    [J]. CENTRAL EUROPEAN JOURNAL OF MATHEMATICS, 2013, 11 (04): : 595 - 596
  • [49] Adjoints of large simulation codes through Automatic Differentiation
    Hascoet, Laurent
    Dauvergne, Benjamin
    [J]. EUROPEAN JOURNAL OF COMPUTATIONAL MECHANICS, 2008, 17 (1-2): : 63 - 86
  • [50] Computing Asymptotic Gains of Large-Scale Interconnections
    Rueffer, Bjoern S.
    Ito, Hiroshi
    Dower, Peter M.
    [J]. 49TH IEEE CONFERENCE ON DECISION AND CONTROL (CDC), 2010, : 7413 - 7418
12345