More Scalability, Less Pain: A Simple Programming Model and Its Implementation for Extreme Computing
|Title||More Scalability, Less Pain: A Simple Programming Model and Its Implementation for Extreme Computing|
|Publication Type||Journal Article|
|Year of Publication||2010|
|Authors||Lusk, EL, Pieper, SC, Butler, RM|
A critical issue for the future of high‐performance computing is the programming model to use on next‐generation architectures. Here we describe an approach to programming very large machines that combines simplifying the programming model with a scalable library implementation that we find promising. Our presentation takes the form of a case study in nuclear physics. The chosen application addresses fundamental issues in the origins of our universe, while the library developed to enable this application on the largest computers may have applications beyond this one. The work here relies heavily on DOE�s SciDAC program for funding both the physics and computer science work and its INCITE program to provide the computer time to obtain the results.