Software Testing for Scientific Computing in HPC (SC Website Link)
Held as part of SC23 in Denver, Colorado
Wednesday, November 15th, 12:15-1:15 PM MT
Room 605, Colorado Convention Center
Attendees: 80+ in person; 5+ online
Abstract
Effective software testing plays a critical role in guaranteeing the performance, correctness, and reproducibility of applications and software. When it comes to testing high-performance computing (HPC) software and applications, unique requirements arise due to factors such as massive parallelism, concurrency and heterogeneity, the scale of target platforms, lack of oracles, and application-specific verification and validation techniques. In this BoF session, we aim to foster insightful discussions among a panel of expert speakers and the audience, focusing on methodologies and challenges in HPC software testing, and deepen our understanding in this crucial part of HPC software development.
Format
This Birds of a Feather consisted of short presentations from experts in the field of software testing along with questions from the audience related to software testing topics.
Panelists
Sunita Chandrasekaran – University of Delaware
Talk title: Software Testing (for Directives)
Sunita Chandrasekaran is an Associate Professor with the Department of Computer and Information Sciences (CIS) at the University of Delaware, USA. She is also a computational scientist with Brookhaven National Laboratory. Her research interests span High Performance Computing including exascale computing, parallel programming models, performance analysis, machine learning, and developing research scientific software for large scale real-world applications targeting upcoming systems. Her work has been recognized with a 2016 IEEE Technical Consortium on High Performance Computing (TCHPC) for Early Career Researchers award among other awards. Sunita holds a PhD in computer science engineering from Nanyang Technological University, Singapore and a bachelor’s degree in electrical and electronics from Anna University, India.
Mark Gates – University of Tennessee
Talk title: Testing the SLATE Linear Algebra Library
Mark Gates is a research scientist at the Innovative Computing Laboratory at the University of Tennessee, Knoxville. His research is in the solution of linear algebra problems on modern computers with multi-core and GPU architectures, as part of the SLATE, PLASMA, and MAGMA projects. He received his Ph.D. in computer science from the University of Illinois at Urbana-Champaign, where his research investigated the parallel implementation of digital volume correlation, a method used in mechanical engineering.
Sean McGovern – NexGen Analytics
Talk title: Testing with Trilinos
Sean is a Principal Scientist and the Director of Scientific Computing, leading the Chapel Hill, NC office of NexGen Analytics. Sean was trained in physical science (Physics, MS, Virginia) and computational science at the Forschungszentrum Juelich (Geophysics, PhD, Bonn). Sean worked in applied mathematics as a Post-Doc at INRIA/Universite Cote D’Azur before beginning to collaborate with NGA in 2022.
Manish Motwani – Oregon State University
Talk title: Automatically Testing Correctness of HPC Scientific Applications
Manish is an assistant professor at Oregon State University. His research involves analyzing large software repositories to learn interesting phenomena in software development and maintenance, and to use that knowledge to design novel automation techniques, such as testing and program repair.
Shahzeb Siddiqui – LBNL/NERSC
Talk title: Automated Software Testing of Spack/E4S with buildtest
Shahzeb Siddiqui is a HPC Consultant/Software Integration Specialist at Lawrence Berkeley National Laboratory/NERSC.
Organizers
Keita Teranishi, Oak Ridge National Laboratory
Keita Teranishi is a senior computer scientist and the group leader of the programming systems group with Oak Ridge National Laboratory (ORNL). He has contributed to efforts related to software engineering and performance portability among many other HPC topics including programming languages, macro-network simulation, fault-tolerance, numerical algorithms, and performance tuning. Prior to ORNL, he was a principal member of technical staff at Sandia National Laboratories, and software engineer in the math and scientific libraries group at Cray Inc. Dr. Teranishi received the BS and MS degrees from the University of Tennessee, Knoxville, in 1998 and 2000, respectively, and the PhD degree from Penn State University, in 2004.
Jeffrey Young, Georgia Tech
Dr. Young is a senior research scientist in the School of Computer Science at Georgia Tech and the associate director of Georgia Tech’s new Scientific Software Engineering Center. Dr. Young has been engaged with the scientific software community since 2013 with involvement in Georgia Tech’s Keeneland project, the PIConGPU CAAR effort for Frontier, and Georgia Tech’s HPC student cluster competition team. Young is also the director of Georgia Tech’s novel architecture testbed, the CRNCH Rogues Gallery, and is the managing director of the Arm HPC User Group
Alex Orso, Georgia Tech
Dr. Alex Orso is the director of Georgia Tech’s new Scientific Software Engineering, which is focused on bringing software engineering techniques to a wider scientific audience. His research is in software engineering (SE) with an emphasis on the development of techniques targeted at real-world systems. Orso served on the editorial boards of the major SE journals (TOSEM and TSE) and has been program chair or co-chair for the main SE conferences (ICSE and FSE). He has served on the Advisory Board of Reflective Corp, has been a technical consultant to DARPA, and has received funding for his research from both government agencies, such as DARPA, ONR, and NSF, and industry, such as Facebook, Fujitsu Labs, Google, IBM, and Microsoft. Orso is a Distinguished Member of the ACM and an IEEE Fellow.