LFRic:

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

S. V. Adams, R. W. Ford, M. Hambley, J. M. Hobson, I. Kavcic, C. M. Maynard, T. Melvin, Eike Müller, S. Mullerworth, A. R. Porter, M. Rezney, B. J. Shipway, R. Wong

Research output: Contribution to journalArticle

Abstract

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.

Original languageEnglish
Pages (from-to)383-396
Number of pages14
JournalJournal of Parallel and Distributed Computing
Volume132
Early online date21 May 2019
DOIs
Publication statusPublished - 1 Oct 2019
EventExascale Applications and Software 2018 - Edinburgh, UK United Kingdom
Duration: 17 Apr 201819 Apr 2018
http://www.easc2018.ed.ac.uk/

Fingerprint

Climate models
Scalability

Keywords

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

ASJC Scopus subject areas

  • Software
  • Theoretical Computer Science
  • Hardware and Architecture
  • Computer Networks and Communications
  • Artificial Intelligence

Cite this

LFRic: Meeting the challenges of scalability and performance portability in Weather and Climate models. / Adams, S. V.; Ford, R. W.; Hambley, M.; Hobson, J. M.; Kavcic, I.; Maynard, C. M.; Melvin, T.; Müller, Eike; Mullerworth, S.; Porter, A. R.; Rezney, M.; Shipway, B. J.; Wong, R.

In: Journal of Parallel and Distributed Computing, Vol. 132, 01.10.2019, p. 383-396.

Research output: Contribution to journalArticle

Adams, SV, Ford, RW, Hambley, M, Hobson, JM, Kavcic, I, Maynard, CM, Melvin, T, Müller, E, Mullerworth, S, Porter, AR, Rezney, M, Shipway, BJ & Wong, R 2019, 'LFRic: Meeting the challenges of scalability and performance portability in Weather and Climate models', Journal of Parallel and Distributed Computing, vol. 132, pp. 383-396. https://doi.org/10.1016/j.jpdc.2019.02.007
Adams, S. V. ; Ford, R. W. ; Hambley, M. ; Hobson, J. M. ; Kavcic, I. ; Maynard, C. M. ; Melvin, T. ; Müller, Eike ; Mullerworth, S. ; Porter, A. R. ; Rezney, M. ; Shipway, B. J. ; Wong, R. / LFRic: Meeting the challenges of scalability and performance portability in Weather and Climate models. In: Journal of Parallel and Distributed Computing. 2019 ; Vol. 132. pp. 383-396.
@article{b65e20ab18db4fdca226f1868abd5f5c,
title = "LFRic:: Meeting the challenges of scalability and performance portability in Weather and Climate models",
abstract = "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.",
keywords = "Domain specific language, Exascale, Numerical weather prediction, Separation of concerns",
author = "Adams, {S. V.} and Ford, {R. W.} and M. Hambley and Hobson, {J. M.} and I. Kavcic and Maynard, {C. M.} and T. Melvin and Eike M{\"u}ller and S. Mullerworth and Porter, {A. R.} and M. Rezney and Shipway, {B. J.} and R. Wong",
year = "2019",
month = "10",
day = "1",
doi = "10.1016/j.jpdc.2019.02.007",
language = "English",
volume = "132",
pages = "383--396",
journal = "Journal of Parallel and Distributed Computing",
issn = "0743-7315",
publisher = "Elsevier",

}

TY - JOUR

T1 - LFRic:

T2 - Meeting the challenges of scalability and performance portability in Weather and Climate models

AU - Adams, S. V.

AU - Ford, R. W.

AU - Hambley, M.

AU - Hobson, J. M.

AU - Kavcic, I.

AU - Maynard, C. M.

AU - Melvin, T.

AU - Müller, Eike

AU - Mullerworth, S.

AU - Porter, A. R.

AU - Rezney, M.

AU - Shipway, B. J.

AU - Wong, R.

PY - 2019/10/1

Y1 - 2019/10/1

N2 - 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.

AB - 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.

KW - Domain specific language

KW - Exascale

KW - Numerical weather prediction

KW - Separation of concerns

UR - http://www.scopus.com/inward/record.url?scp=85066413388&partnerID=8YFLogxK

U2 - 10.1016/j.jpdc.2019.02.007

DO - 10.1016/j.jpdc.2019.02.007

M3 - Article

VL - 132

SP - 383

EP - 396

JO - Journal of Parallel and Distributed Computing

JF - Journal of Parallel and Distributed Computing

SN - 0743-7315

ER -