Welcome to the 2015 International Workshop on Software Engineering for High Performance Computing in Science. The submissions covered a wide array of topics from software engineering, scientific computing, and high performance computing. This workshop contains two types of papers: full papers (8 pages) and position papers (4 pages). After a thorough review and discussion process, the program committee and organizing committee accepted six full papers and four position papers for inclusion in the program.
The primary goal of this workshop is to facilitate the interaction among researchers from domains who do not usually have the opportunity to interact. As a result, this workshop always generates interesting and lively discussions. We expect that the papers accepted for this year's program will provide the necessary context and discussion points to once again have interesting and productive discussions. The organizing committee is committed to devoting a significant portion of the workshop time to small and large group discussions. Due to the large amount of time devoted to discussion, we encourage participation from interested researchers who do not have a paper on the program.
Proceeding Downloads
Supporting scientists in re-engineering sequential programs to parallel using model-driven engineering
Developing complex computational-intensive and data-intensive scientific applications requires effective utilization of the computational power of the available computing platforms including grids, clouds, clusters, multi-core and many-core processors, ...
Development of scientific software for HPC architectures using OpenACC: the case of LQCD
- Claudio Bonati,
- Enrico Calore,
- Simone Coscetti,
- Massimo D'Elia,
- Michele Mesiti,
- Francesco Negro,
- Sebastiano Fabio Schifano,
- Raffaele Tripiccione
Many scientific software applications, that solve complex compute- or data-intensive problems, such as large parallel simulations of physics phenomena, increasingly use HPC systems in order to achieve scientifically relevant results. An increasing ...
A scientific function test framework for modular environmental model development: application to the community land model
As environmental models have become more complicated, we need new tools to analyze and validate these models and to facilitate collaboration among field scientists, observation dataset providers, environmental system modelers, and computer scientists. ...
Commit quality in five high performance computing projects
High Performance Computing (HPC) has a long history of software development but relatively little is known about the approaches this community uses to create and maintain software. To close this gap we study the practices of using version control tools ...
Regression testing of GPU/MIC systems for HPCC
Multicore GPU, Intel MIC, and FPGA supplemental parallel processors have become widely implemented in High Performance Computing Clusters (HPCCs). In HPCCs, Computing nodes are assembled with these supplemental processors for specific research ...
SIMPL: a pattern language for writing efficient kernels on GPGPU
Graphics processing units (GPUs) have become an integral part of both High Performance Computing (HPC) and desktop systems. To fully exploit their potential, algorithms should be specifically designed to fit the General Purpose computing on GPU (GPGPU) ...
Computation for genomics knowledge discovery
Knowledge discovery in genomics involves large scale graph processing and inference which is different from high-performance computing in genomics for sequence analysis.
Genomics datasets are becoming increasing large and varied due to advances in ...
Towards an engineering methodology for multi-model scientific simulations
Complex physical phenomena are characterized by sub-systems that continuously interact with each other, and that can be modeled with different computational models. To study such phenomena we need to integrate the heterogeneous computational models of ...
Ensuring an effective user experience when managing and running scientific HPC software
With CPU clock speeds stagnating over the last few years, ongoing advances in computing power and capabilities are being supported through increasing multi- and many-core parallelism. The resulting cost of locally maintaining large-scale computing ...
Using software engineering methodologies to port a scientific code to GPUs: experiences and lessons learned: position paper
This work was carried out as a master's level software engineering project over the course of two semesters. Software engineering methodologies were applied to try to improve the portability, maintainability, and performance of a fusion energy code by ...