Quantcast

Reliability in the Face of Complexity; The Challenge of High-End Scientific Computing

Research paper by Gary Ferland

Indexed on: 07 Oct '02Published on: 07 Oct '02Published in: Astrophysics



Abstract

This conference centers on photoionized plasmas, and the tools necessary to understand them. One of the major goals of meetings held between developers of plasma codes is to identify sources of differences between various codes, and resolve them. Very often these differences are caused by bugs. This contribution outlines some practical aspects of developing bug-free codes, and is based on my experience creating the plasma simulation code Cloudy. Computers present nothing less than a new way to understand the universe. We can do any numerical experiment we can imagine. But machines are growing faster far more quickly than people are getting smarter. Harnessing the power of future machines in a reliable way will present an ever greater intellectual challenge. More powerful machines allow a more complete and complex simulation of physical processes, and so offers a far greater opportunity for errors. Industrial standards of quality control, and the associated code development techniques, can play a major role in developing reliable simulations.