6: Deliver Computing for the Frontiers of Science

Deliver forefront computational and networking capabilities to scientists nationwide that enable them to extend the frontiers of science, answering critical questions that range from the function of living cells to the power of fusion energy.

Other Information:

Executive Summary: Each of the previous goals, and progress in many other areas of science, depends critically on advances in computational modeling and simulation. Crucial problems that we can only hope to address computationally require us to deliver orders of magnitude greater effective computing power than we can deploy today. Detailed Commentary: Computer-based simulation enables us to predict the behavior of complex systems that are beyond the reach of our most powerful experimental probes or our most sophisticated theories. Computational modeling has greatly advanced our understanding of fundamental processes of Nature, such as fluid flow and turbulence or molecular structure and reactivity. Through modeling and simulation, we will be able to explore the interior of stars and learn how protein machines work inside living cells. We can design novel catalysts and high-efficiency engines. Computational science is increasingly central to progress at the frontiers of almost every scientific discipline and to our most challenging feats of engineering. The science of the future demands that we advance beyond our current computational abilities. Accordingly, we must address the following challenges: • What new mathematics are required to effectively model systems such as the Earth’s climate or the behavior of living cells that involve processes taking place on vastly different time and/or length scales? • Which computational architectures and platforms will deliver the most benefit for the science of today and the science of the future? • What advances in computer science and algorithms are needed to increase the efficiency with which supercomputers solve problems for the Office of Science? • What operating systems, data management, analysis, model development, and other tools are required to make effective use of future-generation supercomputers? • Is it possible to overcome the geographical distances that often hinder science by making all scientific resources readily available to scientists, regardless of whether they are at a university, national laboratory, or industrial setting? The Office of Science will deliver models, tools, and computing platforms to dramatically increase the effective computational capability available for scientific discovery in fusion, nanoscience, highenergy and nuclear physics, climate and environmental science, and biology. We will develop new mathematics and computational methods for modeling complex systems; work with the scientific community and vendors to develop computing architectures tailored to simulation and modeling; develop improved networking resources; and support interdisciplinary teams of scientists, mathematicians, and computer scientists to build sophisticated computational models that fully exploit these capabilities. Our role complements and builds on the National Nuclear Security Administration’s Accelerated Strategic Computing Initiative, delivering forefront modeling capabilities for stockpile stewardship, the basic computer science and mathematics research programs conducted by the National Science Foundation, and mission-focused programs of other agencies. As an integral part of this Strategic Plan, and in Facilities for the Future of Science: A Twenty-Year Outlook, we have identified the need for three future facilities to realize our Advanced Scientific Computing Research vision and to meet the science challenges described in the following pages. All three of the facilities are near-term priorities: the UltraScale Scientific Computing Capability (USSCC), the Energy Sciences Network (ESnet) Upgrade, and the National Energy Research Scientific Computing Center (NERSC) Upgrade. The USSCC, located at multiple sites, will increase by a factor of 100 the computing capability available to support open (as opposed to classified) scientific research—reducing from years to days the time required to simulate complex systems, such as the chemistry of a combustion engine, or weather and climate—and providing much finer resolution. The ESnet upgrade will enhance the network services available to support Office of Science researchers and laboratories and maintain their access to all major DOE research facilities and computing resources, as well as fast interconnections to more than 100 other networks. The NERSC upgrade will ensure that DOE’s premier scientific computing facility for unclassified research continues to provide high-performance computing resources to support the requirements of scientific discovery. All three facilities are included in our Advanced Scientific Computing Research Strategic Timeline at the end of this chapter and in the facilities chart in Chapter 7 (page 93), and they are discussed in detail in the Twenty-Year Outlook. Our Timeline and Indicators of Success: Our commitment to the future and to the realization of Goal 6: Deliver Computing for the Frontiers of Science is not only reflected in our strategies, but also in our Key Indicators of Success, below, and our Strategic Timeline for Advanced Scientific Computing Research (ASCR), at the end of this chapter. The ASCR Strategic Timeline charts a collection of important, illustrative milestones, representing planned progress within each strategy. These milestones, while subject to the rapid pace of change and uncertainties that belie all science programs, reflect our latest perspectives on the future— what we hope to accomplish and when we hope to accomplish it— over the next 20 years and beyond. Following the science milestones, toward the bottom of the timeline, we have identified the required major new facilities. These facilities, described in greater detail in the DOE Office of Science companion report, Facilities for the Future of Science: A Twenty-Year Outlook, reflect time-sequencing that is based on the general priority of the facility, as well as critical-path relationships to research and corresponding science milestones. Additionally, the Office of Science has identified Key Indicators of Success, designed to gauge our overall progress toward achieving Goal 6. These select indicators, identified below, are representative long-term measures against which progress can be evaluated over time. The specific features and parameters of these indicators, as well as definitions of success, can be found on the web at www.science.doe.gov/ measures. Key Indicators of Success: • Progress toward developing the mathematics, algorithms, and software that enable effective scientifically critical models of complex systems, including highly nonlinear or uncertain phenomena, or processes that interact on vastly different scales or contain both discrete and continuous elements. • Progress toward developing, through the Genomics: GTL partnership with the Biological and Environmental Research program, the computational science capability to model a complete microbe and a simple microbial community.

Objective(s):

6.1: Complex Systems

6.2: Computers, Collaboratory Software, and Computational Models

6.3: Supercomputing Architectures

6.4: Computing Resources and Network Infrastructure

Content