The 2015 International Workshop on Software Engineering for High Performance Computing in Science

NEWS: The best papers from the workshop will be invited to submit expanded versions to Computing in Science & Engineering

Shared Google Document

Overview

This workshop is concerned with identifying and understanding the unique aspects of Software Engineering (SE) for the development of scientific software that exploits High Performance Computing (HPC) architectures. This class of software applications includes software developed to support various scientific endeavors which would be difficult or impossible to perform experimentally. This type of software development has not received enough attention from the SE community. Specifically, we are interested in:

• Scientific software applications that solve complex software- or data-intensive research problems. These applications range from large, parallel models/simulations of the physical world using HPC systems to smaller scale simulations developed by a single scientist or engineer on a desktop machine or on a small cluster, which may be later scaled to an HPC environment.
• Applications that support scientific research and experiments. Such applications include, but are not limited to, systems for managing and/or manipulating large amounts of data and systems that provide HPC infrastructure for research applications.
• The process for building, reusing, and publishing HPC software and the data used in scientific experiments or engineering innovations. Among others, these process include agile approaches, open source/open data issues, testing HPC software, and managing software or data repositories for publishing goals.

Despite its importance, the development of scientific HPC software historically has attracted less attention from the SE community than other subdomains have. Indeed, the development of this type of software is significantly different than the development of business information systems, from which many of the SE best practices, tools and techniques have been drawn. Therefore, in order to identify and develop appropriate methods, tools and techniques for scientific HPC software, members of the SE community must interact with members of the HPC and science communities to understand the differences and determine the most appropriate SE tools, methods and techniques. In addition, we hope to identify aspects of SE practice that are relevant for the education of future developers in various research domains, which can be provided as suggestions for inclusion in Software Engineering and Computer Science curricula.

In addition to presentation and discussion of the accepted papers, significant time during the workshop will be devoted to large and small group discussions among the participants to identify important research questions at the intersection of SE and HPC CSE that are in need of additional study.

In previous workshops in this area, the discussion has focused on topics including:

• The demanding characteristics of research software that affect its development choices;
• The appropriate context dimensions to describe research software;
• The major software quality goals for research software to run on HPC architectures;
• Crossing the communication chasm between SE and Science when experiments are software or data-intensive;
• Effectively involving scientists in software development and training;
• Measuring the impact of SE on scientific productivity;
• SE tools and methods needed by the scientific community; and
• How to effectively test scientific HPC software.

In this workshop, the discussions will build on these themes and identify new topics of interest. For more information contact Jeffrey Carver.

---

Last Updated on December 11, 2014 by Jeffrey Carver