The Performance Engineering Institute (PERI) originally proposed a tiger team activity as a mechanism to target significant effort optimizing key Office of Science applications, a model that was successfully realized with the assistance of two JOULE metric teams. However, the Office of Science requested a new focus beginning in 2008: assistance in forming its ten year facilities plan. To meet this request, PERI formed the Architecture Tiger Team, which is modeling the performance of key science applications on future architectures, with S3D, FLASH and GTC chosen as the first application targets. In this activity, we have measured the performance of these applications on current systems in order to understand their baseline performance and to ensure that our modeling activity focuses on the right versions and inputs of the applications. We have applied a variety of modeling techniques to anticipate the performance of these applications on a range of anticipated systems. While our initial findings predict that Office of Science applications will continue to perform well on future machines from major hardware vendors, we have also encountered several areas in which we must extend our modeling techniques in order to fulfill our mission accurately and completely. In addition, we anticipate that models of a wider range of applications will reveal critical differences between expected future systems, thus providing guidance for future Office of Science procurement decisions, and will enable DOE applications to exploit machines in future facilities fully. 1 1. Introduction Sustained performance improvements are integral to the DOE Office of Science SciDAC program's mission to advance large-scale scientific modeling and simulation. Simulation is a key investigative technique for disciplines where experimentation is expensive, dangerous, or impossible. Increased performance can enable faster simulations and more timely predictions, or it can be used to increase the accuracy of existing physical models, enabling more predictive simulations. Research enabled by the SciDAC program will have far-reaching effects in fields such as basic energy, biology, environmental science, fusion energy, and high-energy physics. The Performance Engineering Research Institute (PERI) tiger team activity targets critical SciDAC performance needs. The original intent was for each tiger team to focus the efforts of several PERI researchers on improving performance of an Office of Science application, with the application selected based on Office of Science mission objectives and application readiness for the focused effort. Thus, each tiger team was envisioned as a relatively short-term activity (six months to at most one year). In 2007, our tiger teams had a positive impact on two key DOE applications participating in the JOULE metric. We improved the performance of a turbulent combustion code (S3D [1]) on Oak Ridge's Cray XT5 Jaguar system by 13%. Similarly, we improved the performance of the Gyrokinetic Toroidal Code (GTC) [2, 3] by 10% on Ja...
Abstract. The Performance Engineering Institute (PERI) originally proposed a tiger team activity as a mechanism to target significant effort optimizing key Office of Science applications, a model that was successfully realized with the assistance of two JOULE metric teams. However, the Office of Science requested a new focus beginning in 2008: assistance in forming its ten year facilities plan. To meet this request, PERI formed the Architecture Tiger Team, which is modeling the performance of key science applications on future architectures, with S3D, FLASH and GTC chosen as the first application targets. In this activity, we have measured the performance of these applications on current systems in order to understand their baseline performance and to ensure that our modeling activity focuses on the right versions and inputs of the applications. We have applied a variety of modeling techniques to anticipate the performance of these applications on a range of anticipated systems. While our initial findings predict that Office of Science applications will continue to perform well on future machines from major hardware vendors, we have also encountered several areas in which we must extend our modeling techniques in order to fulfill our mission accurately and completely. In addition, we anticipate that models of a wider range of applications will reveal critical differences between expected future systems, thus providing guidance for future Office of Science procurement decisions, and will enable DOE applications to exploit machines in future facilities fully. IntroductionSustained performance improvements are integral to the DOE Office of Science SciDAC program's mission to advance large-scale scientific modeling and simulation. Simulation is a key investigative technique for disciplines where experimentation is expensive, dangerous, or impossible. Increased performance can enable faster simulations and more timely predictions, or it can be used to increase the accuracy of existing physical models, enabling more predictive simulations. Research enabled by the SciDAC program will have far-reaching effects in fields such as basic energy, biology, environmental science, fusion energy, and high-energy physics. The Performance Engineering Research Institute (PERI) tiger team activity targets critical SciDAC performance needs. The original intent was for each tiger team to focus the efforts of several PERI researchers on improving performance of an Office of Science application, with the application selected based on Office of Science mission objectives and application readiness for the focused effort. Thus, each tiger team was envisioned as a relatively short-term activity (six months to at most one year). In 2007, our tiger teams had a positive impact on two key DOE applications participating in the JOULE metric. We improved the performance of a turbulent combustion code (S3D [1]) on Oak Ridge's Cray XT5 Jaguar system by 13%. Similarly, we improved the performance of the Gyrokinetic Toroidal Code (GTC) [2,3] by 10% o...
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.