Preprints
https://doi.org/10.5194/gmd-2021-265
https://doi.org/10.5194/gmd-2021-265

Submitted as: methods for assessment of models 26 Aug 2021

Submitted as: methods for assessment of models | 26 Aug 2021

Review status: this preprint is currently under review for the journal GMD.

A Scalability Study of the Ice-sheet and Sea-level System Model (ISSM, Version 4.18)

Yannic Fischler1, Martin Rückamp2, Christian Bischof1, Vadym Aizinger3, Mathieu Morlighem4,5, and Angelika Humbert2,6 Yannic Fischler et al.
  • 1Department of Computer Science, Technical University Darmstadt, Darmstadt, Hesse, Germany
  • 2Alfred-Wegener-Institut, Helmholtz-Zentrum für Polar- und Meeresforschung, Bremerhaven, Bremen, Germany
  • 3Chair of Scientific Computing, University of Bayreuth, Bayreuth, Bavaria, Germany
  • 4Department of Earth Sciences, Dartmouth College, Hanover, United States of America
  • 5Department of Earth System Science, University of California Irvine, United States of America
  • 6Faculty of Geosciences, University of Bremen, Bremen, Germany

Abstract. Accurately modeling the contribution of Greenland and Antarctica to sea level rise requires to solve partial differential equations at a high spatial resolution. It is important to test the scalability of existing ice sheet models in order to assess whether they are ready to take advantage of new cluster architectures. In this paper, we discuss the overall scaling of the Ice-sheet and Sea-level System Model (ISSM) applied to the Greenland ice sheet. The model setup used as benchmark problem comprises a variety of modules with different levels of complexity and computational demands. The core builds the so-called stress balance module, which uses the higher-order approximation (or Blatter-Pattyn) of the Stokes equations and a mesh of linear prismatic finite elements to compute the ice flow. We develop a detailed user-oriented, yet low-overhead performance instrumentation tailored to the requirements of earth system models and run scaling tests up to 6 144 MPI processes. The results show that the computation of the Greenland model scales overall well up to 3 072 MPI processes, but is eventually slowed down by matrix assembly, the output handling, and lower-dimensional problems that employ lower numbers of unknowns per MPI process. We also discuss improvements of the scaling and identify further improvements needed for climate research. The instrumented version of ISSM, thus, not only identifies potential performance bottlenecks that were not present at lower core counts but also provides the capability to continually monitor the performance of ISSM code basis. This is of long-term significance as the overall performance of ISSM model depends on the subtle interplay between algorithms, their implementation, underlying libraries, compilers, run-time systems and hardware characteristics, all of which are in a constant state of flux.

Yannic Fischler et al.

Status: open (until 21 Oct 2021)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse

Yannic Fischler et al.

Data sets

ISSM Profiling Data Yannic Fischler https://doi.org/10.48328/tudatalib-612

Greenland Setup Martin Rückamp https://doi.org/10.48328/tudatalib-614

Model code and software

modified ISSM v4.18 modifications by Yannic Fischler and Martin Rückamp https://doi.org/10.48328/tudatalib-613

Yannic Fischler et al.

Viewed

Total article views: 250 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
194 53 3 250 2 3
  • HTML: 194
  • PDF: 53
  • XML: 3
  • Total: 250
  • BibTeX: 2
  • EndNote: 3
Views and downloads (calculated since 26 Aug 2021)
Cumulative views and downloads (calculated since 26 Aug 2021)

Viewed (geographical distribution)

Total article views: 157 (including HTML, PDF, and XML) Thereof 157 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 28 Sep 2021
Download
Short summary
Ice sheet models are used to simulate the changes of ice sheets in future, but are currently often run in coarse resolution and/or with neglecting important physics to make them affordable in terms of computational costs. We conducted a study simulating the Greenland Ice Sheet in high resolution and adequate physics to test where the ice sheet code ISSM is using most time and what could be done to improve its performance for future computer architectures that allow massive parallel computing.