Collaborative Research: DOE/NSF Workshop on Correctness in Scientific Computing

Rutgers University New Brunswick, New Brunswick NJ

Investigators

Abstract

The DOE/NSF Workshop on Correctness in Scientific Computing (CSC) will be held on June 17-18, 2023, as part of the 44th ACM SIGPLAN Conference on Programming Language Design and Implementation (PLDI 2023). PLDI 2023 will be one of 14 leading computer science conferences taking place simultaneously as part of the ACM Federated Computing Research Conference (FCRC) in Orlando, Florida. Several stakeholder communities, for example, theory, programming languages design and implementation, computer architecture, high-performance parallel and distributed computing, and supercomputing, will be in attendance: thus, the location of this workshop at FCRC. The website for the workshop is https://pldi23.sigplan.org/home/csc-2023 While improving performance of scientific computing systems continues to be important, there is growing awareness that ensuring correctness of these systems is a major challenge. To address this challenge and to provide novel ideas and perspectives, the workshop brings together experts from among the areas of numerical methods, scientific computing, formal methods, high-performance computing, compilation, testing, static and dynamic program analysis, computer architecture, and error analysis. Beyond discussing latest advances and technical accomplishments, the workshop will inform DOE and NSF of important and groundbreaking future research directions pertaining to correctness in scientific computing. The workshop's impacts include broadening participants' perspectives and providing a venue to form new collaborations. Furthermore, a post-workshop report describing the most important and promising research directions will be made available to the broader research community, to inspire research in those directions. Ultimately, such research will lead to a new set of robust solutions to correctness problems in scientific computing, leading to higher quality software and computing systems, and greater confidence that these systems are producing correct results. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.