Performance Optimization of Parallel Applications in Diverse On-Demand Development Teams

Iliev, Hristo; Halver, René; Müller, Matthias S.; Terboven, Christian; Hermanns, Marc-André; Göbbert, Jens Henrik; Mohr, Bernd

doi:10.1007/978-3-319-53862-4_16

Contribution to a conference proceedings/Contribution to a book

FZJ-2017-01976

Performance Optimization of Parallel Applications in Diverse On-Demand Development Teams

Iliev, H. (Corresponding author) ; Hermanns, M.-A.FZJ* ; Göbbert, J. H.FZJ* ; Halver, R.FZJ* ; Terboven, C. ; Mohr, B.FZJ* ; Müller, M. S.

2017
Springer International Publishing Cham
ISBN: 978-3-319-53861-7 (print), 978-3-319-53862-4 (electronic)

High-Performance Scientific Computing / Di Napoli, Edoardo (Editor) ; Cham : Springer International Publishing, 2017, Chapter 16 ; ISSN: 0302-9743=1611-3349 ; ISBN: 978-3-319-53861-7=978-3-319-53862-4 ; doi:10.1007/978-3-319-53862-4
First JARA-HPC Symposium 2016, JHPCS'16, Aachen, Germany, 4 Oct 2016 - 5 Oct 2016 Cham : Springer International Publishing, Lecture Notes in Computer Science 10164, 187 - 199 (2017) [10.1007/978-3-319-53862-4_16]

This record in other databases:

Please use a persistent id in citations: http://hdl.handle.net/2128/13902 doi:10.1007/978-3-319-53862-4_16

Abstract: Current supercomputing platforms and scientific application codes have grown rapidly in complexity over the past years. Multi-scale, multi-domain simulations on one hand and deep hierarchies in large-scale computing platforms on the other make it exceedingly harder to map the former onto the latter and fully exploit the available computational power. The complexity of the software and hardware components involved calls for in-depth expertise that can only be met by diversity in the application development teams. With its model of simulation labs and cross-sectional groups, JARA-HPC enables such diverse teams to form on demand to solve concrete development problems. This work showcases the effectiveness of this model with two application case studies involving the JARA-HPC cross-sectional group “Parallel Efficiency” and simulation labs and domain-specific development teams. For one application, we show the results of a completed optimization and the estimated financial impact of the combined efforts. For the other application, we present results from an ongoing engagement, where we show how an on-demand team investigates the behavior of dynamic load balancing schemes for an MD particle simulation, leading to a better overall understanding of the application and revealing targets for further investigation.

Classification: