Visit additional Tabor Communication Publications
November 25, 2005
In June 2004, the President's Information Technology Advisory Committee (PITAC) was charged by John Marburger, the President's Science Advisory, to respond to seven questions regarding the state of computational science. Following over a year of hearings and deliberations, the committee released its report, entitled Computational Science: Ensuring America's Competitiveness, in June 2005. What follows are some of my personal perspectives on computational science, shaped by the committee experience. Any wild eyed, crazy ideas should be attributed to me, not to the committee.
Based on community input and extensive discussions, the PITAC computational science report included the following principal finding and recommendation.
Computational science is now indispensable to the solution of complex problems in every sector, from traditional science and engineering domains to such key areas as national security, public health, and economic innovation. Advances in computing and connectivity make it possible to develop computational models and capture and analyze unprecedented amounts of experimental and observational data to address problems previously deemed intractable or beyond imagination. Yet, despite the great opportunities and needs, universities and the Federal government have not effectively recognized the strategic significance of computational science in either their organizational structures or their research and educational planning. These inadequacies compromise U.S. scientific leadership, economic competitiveness, and national security.
Succinctly, the principal finding highlights the emergence of computational science as the third pillar of scientific discovery, as a complement to theory and experiment. It also highlights the critical importance of computational science to innovation, security and scientific discovery, together with our failure to embrace computational science as a strategic, rather than a tactical capability. In many ways, computational science has been everyone's "second priority," rather than the unifying capability it could be.
Universities and the Federal government's R&D agencies must make coordinated, fundamental, structural changes that affirm the integral role of computational science in addressing the 21st century's most important problems, which are predominantly multidisciplinary, multi-agency, multisector, and collaborative. To initiate the required transformation, the Federal government, in partnership with academia and industry, must also create and execute a multi-decade roadmap directing coordinated advances in computational science and its applications in science and engineering disciplines.
The principal recommendation emphasizes the silos and stovepipes (choose your favorite analogy) that separate disciplinary domains within computational science. There was widespread consensus from both those who testified and those on the committee that solving many of the most important problems of the 21st century will require integration of skills from diverse groups. The group also felt deeply that current organizational structures in academia and government placed limits on interdisciplinary education and research.
Based on this recognition, the committee's principal recommendation was to create a long-term, regularly updated strategic roadmap of technologies (i.e., software, data management, architectures and systems, and programming and tools), application needs and their interplay. The long term, strategic aspect of this recommendation cannot be over-estimated. Many of our most important computational science challenges cannot be solved in 1-3 years. Nor is a series of three year plans the same as a 10-15 year plan.
Substantial, sustained investment, driven by multi-agency collaboration, is the only approach that will allow us to escape from our current technology quandary-high-performance computing systems that are based on fragile software and an excessive emphasis on peak performance, rather than sustained performance on important applications. Simply put, today's computational science ecosystem is unbalanced, with a software and hardware base that is inadequate to keep pace with and support evolving application needs. By starving research in enabling software and hardware, the imbalance forces researchers to build atop crumbling and inadequate foundations. The result is greatly diminished productivity for both researchers and computing systems.
Similarly, we must embrace the data explosion from large-scale instruments and ubiquitous, microscale sensors-the personal petabyte is in sight! Given the strategic significance of this scientific trove, the Federal government must provide long-term support for computational science community data repositories. HPC cannot remain synonymous with computing, but must be defined broadly to include distributed sensors and storage.
In the 19th and 20th centuries, proximity to transportation systems (navigable rivers, seaports, railheads, and airports) was critical to success. Cities grew and developed around such transportation systems, providing jobs and social services. In today's information economy, high-speed networking, data archives and computing systems play a similar role, connecting intellectual talent across geographic barriers via virtual organizations (VOs)-teams drawn from multiple organizations, with diverse skills and access to wide ranging resources, that can coordinate and leverage intellectual talent. Two examples serve to illustrate both the challenges and the opportunities that could accrue from visionary application of computational science.
Hurricane Katrina drove home the centrality of VOs. In computational science terms, a rapid response VO would include integrated hurricane, storm surge, tornado spawning, environmental, transportation, communication and human dynamics models, together with the experts needed to analyze model outputs and shape public policy for evacuation, remediation and recovery. Computationally, solving such a complex problem requires real-time data fusion from wide arrays of distributed sensors, large and small; coupled, computational intense environmental models; and social behavior models. There are thousands of such 21st century problems, each awaiting application of computational science tools and techniques.
The fusion of knowledge from genomics, protein structure, enzyme function and pathway and regulatory models to create systemic models of organelles, cells and organisms and their relation to the environment is one of the great biological challenges of the 21st century. By combining information from experiments, data gleaned from mining large-scale archives (e.g., genomic, proteomic, structural and other data), and large-scale biological simulations and computational models, we can gain insights into function and behavior-understanding life in a deep way. The time is near to mount a multidisciplinary effort to create artificial life, a computational counterpart to Craig Venter's minimal genome project. Such an effort would combine engineering, genomics, proteomics and systems biology expertise, with profound implications for medicine and deep insights into biology.
The computational science opportunities have never been greater. It is time to act with vision and sustained commitment.
The PITAC report on computational science can be downloaded from www.nitrd.gov. Paper copies of the report can be requested there as well.
This article originally appeared in CTWatch Quarterly, Volume 1, Number 4, November 2005. To view the entire issue visit http://www.ctwatch.org/quarterly/.
Jun 18, 2013 |
The world's largest supercomputers, like Tianhe-2, are great at traditional, compute-intensive HPC workloads, such as simulating atomic decay or modeling tornados. But data-intensive applications--such as mining big data sets for connections--is a different sort of workload, and runs best on a different sort of computer.
Jun 18, 2013 |
Researchers are finding innovative uses for Gordon, the 285 teraflop supercomputer housed at the San Diego Supercomputer Center (SDSC) that has a unique Flash-based storage system. Since going online, researchers have put the incredibly fast I/O to use on a wide variety of workloads, ranging from chemistry to political science.
Jun 17, 2013 |
The advent of low-power mobile processors and cloud delivery models is changing the economics of computing. But just as an economy car is good at different things than a full size truck, an HPC workload still has certain computing demands that neither the fastest smartphone nor the most elastic cloud cluster can fulfill.
Jun 14, 2013 |
For all the progress we've made in IT over the last 50 years, there's one area of life that has steadfastly eluded the grasp of computers: understanding human language. Now, researchers at the Texas Advanced Computing Center (TACC) are utilizing a Hadoop cluster on its Longhorn supercomputer to move the state of the art of language processing a little bit further.
Jun 13, 2013 |
Titan, the Cray XK7 at the Oak Ridge National Lab that debuted last fall as the fastest supercomputer in the world with 17.59 petaflops of sustained computing power, will rely on its previous LINPACK test for the upcoming edition of the Top 500 list.
05/10/2013 | Cleversafe, Cray, DDN, NetApp, & Panasas | From Wall Street to Hollywood, drug discovery to homeland security, companies and organizations of all sizes and stripes are coming face to face with the challenges – and opportunities – afforded by Big Data. Before anyone can utilize these extraordinary data repositories, however, they must first harness and manage their data stores, and do so utilizing technologies that underscore affordability, security, and scalability.
04/15/2013 | Bull | “50% of HPC users say their largest jobs scale to 120 cores or less.” How about yours? Are your codes ready to take advantage of today’s and tomorrow’s ultra-parallel HPC systems? Download this White Paper by Analysts Intersect360 Research to see what Bull and Intel’s Center for Excellence in Parallel Programming can do for your codes.
Join HPCwire Editor Nicole Hemsoth and Dr. David Bader from Georgia Tech as they take center stage on opening night at Atlanta's first Big Data Kick Off Week, filmed in front of a live audience. Nicole and David look at the evolution of HPC, today's big data challenges, discuss real world solutions, and reveal their predictions. Exactly what does the future holds for HPC?
Join our webinar to learn how IT managers can migrate to a more resilient, flexible and scalable solution that grows with the data center. Mellanox VMS is future-proof, efficient and brings significant CAPEX and OPEX savings. The VMS is available today.