LFRic: Meeting the challenges of scalability and performance portability in Weather and Climate models

被引：39

作者：

Adams, S. V. ^{[1
]}

Ford, R. W. ^{[2
]}

Hambley, M. ^{[1
]}

Hobson, J. M. ^{[1
]}

Kavcic, I. ^{[1
]}

Maynard, C. M. ^{[1
,3
]}

Melvin, T. ^{[1
]}

Mueller, E. H. ^{[4
]}

Mullerworth, S. ^{[1
]}

Porter, A. R. ^{[2
]}

Rezny, M. ^{[5
]}

Shipway, B. J. ^{[1
]}

Wong, R. ^{[1
]}

机构：

[1] Met Off, Exeter, Devon, England

[2] STFC Hartree Ctr, Daresbury Lab, Warrington, Cheshire, England

[3] Univ Reading, Dept Comp Sci, Reading, Berks, England

[4] Univ Bath, Dept Math, Bath, Avon, England

[5] Monash Univ, Melbourne, Vic, Australia

来源：

JOURNAL OF PARALLEL AND DISTRIBUTED COMPUTING | 2019年 / 132卷

关键词：

Separation of concerns; Domain specific language; Exascale; Numerical weather prediction;

D O I：

10.1016/j.jpdc.2019.02.007

中图分类号：

TP301 [理论、方法];

学科分类号：

081202 ;

摘要：

This paper describes LFRic: the new weather and climate modelling system being developed by the UK Met Office to replace the existing Unified Model in preparation for exascale computing in the 2020s. LFRic uses the GungHo dynamical core and runs on a semi-structured cubed-sphere mesh. The design of the supporting infrastructure follows object-oriented principles to facilitate modularity and the use of external libraries where possible. In particular, a 'separation of concerns' between the science code and parallel code is imposed to promote performance portability. An application called PSyclone, developed at the STFC Hartree centre, can generate the parallel code enabling deployment of a single source science code onto different machine architectures. This paper provides an overview of the scientific requirement, the design of the software infrastructure, and examples of PSyclone usage. Preliminary performance results show strong scaling and an indication that hybrid MPI/OpenMP performs better than pure MPI. Crown Copyright (C) 2019 Published by Elsevier Inc. All rights reserved.

引用

页码：383 / 396

页数：14

共 42 条

[1] Adams S. V., 2018, PARALLEL COMPUTING I, P485
[2] Anders Logg G.W., 2012, Lecture Notes in Computational Science and Engineering, V84, P731, DOI [10.1007/978-3-642-23099-8, DOI 10.1007/978-3-642-23099-8]
[3] Arakawa A., 1977, METHODS COMPUTATIONA, V17, P173, DOI [DOI 10.1016/B978-0-12-460817-7.50009-4, https://doi.org/10.1016/B978-0-12-460817-7.50009-4]
[4] Balay S, 1997, MODERN SOFTWARE TOOLS FOR SCIENTIFIC COMPUTING, P163
[5] Balay Satish., PETSc Web page
[6] Bertolli C, 2012, LECT NOTES COMPUT SC, V7155, P191, DOI 10.1007/978-3-642-29737-3_22
[7] Blatt M, 2007, LECT NOTES COMPUT SC, V4699, P666
[8] Boffi D., 2013, MIXED FINITE ELEMENT, V44
[9] Analysis of tensor product multigrid
Börm, S
Hiptmair, R
[J]. NUMERICAL ALGORITHMS, 2001, 26 (03) : 219 - 234
[10] CHARNEY JG, 1953, J METEOROL, V10, P71, DOI 10.1175/1520-0469(1953)010<0071:NIOTQG>2.0.CO

← 1 2 3 4 5 →