Experiences in Developing HPC Software with Portable Efficiency

Daniel Friedrich Harlacher (), Harald Klimach () and Sabine Roller ()
Additional contact information
Daniel Friedrich Harlacher: University of Siegen, Simulation Techniques and Scientific Computing
Harald Klimach: University of Siegen, Simulation Techniques and Scientific Computing
Sabine Roller: University of Siegen, Simulation Techniques and Scientific Computing

A chapter in Sustained Simulation Performance 2014, 2015, pp 165-172 from Springer

Abstract: Abstract This contribution is concerned with software engineering for HPC applications in the field of engineering with a focus on CFD. Our goals are general usability, maintainability, portability and of course high performance and high scalability. Usability is a key property to any application as it sets the bounds on the efficiency the user can work with it. Therefore, we focus on this aspect and show how it is entangled with the other goals.Engineering deals with the design of technical devices and their fine tuning. Increasingly efficient devices require highly sophisticated models. This does not only entail a high resolution but also multiple physical phenomena, complex geometries and different length and time scales to be considered. Simulation software in the field of engineering needs to address these requirements, including high performance to handle the large computational demands. This results in sophisticated applications, which are deployed in an increasingly complex machine environment and need to combine fast execution with involved implementations for large models.We share in this article experiences we made in the development of a general mesh based tool chain for engineering applications named APES and its deployment on HPC machines. Specifically we address on the one hand some software engineering concepts, how they are applied, where they can be neglected and which of them we deem especially important in the HPC context. On the other hand we point to shortcomings in the machine installations that require workarounds and break software engineering principles.

Keywords: Multiple Physical Phenomena; Sustained High Performance; Improve Program Execution; Complex Simulation Software; Compile-time Configuration (search for similar items in EconPapers)
Date: 2015
References: Add references at CitEc
Citations:

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:sprchp:978-3-319-10626-7_14

Ordering information: This item can be ordered from
http://www.springer.com/9783319106267

DOI: 10.1007/978-3-319-10626-7_14

Access Statistics for this chapter

More chapters in Springer Books from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().