It takes a village to perform a petascale computation—domain scientists, appliedmathematicians, computer scientists, computer system vendors, program managers, andsupport staff—and the village was assembled during 24–28 June 2007 in Boston's WestinCopley Place for the third annual Scientific Discovery through Advanced Computing(SciDAC) 2007 Conference. Over 300 registered participants networked around 76posters, focused on achievements and challenges in 36 plenary talks, and brainstormed intwo panels. In addition, with an eye to spreading the vision for simulation at thepetascale and to growing the workforce, 115 participants—mostly doctoral students andpost-docs complementary to the conferees—were gathered on 29 June 2007 inclassrooms of the Massachusetts Institute of Technology for a full day of tutorials on theuse of SciDAC software. Eleven SciDAC-sponsored research groups presented theirsoftware at an introductory level, in both lecture and hands-on formats that included liveruns on a local BlueGene/L.
Computation has always been about garnering insight into the behavior of systems toocomplex to explore satisfactorily by theoretical means alone. Today, however,computation is about much more: scientists and decision makers expect quantitativelyreliable predictions from simulations ranging in scale from that of the Earth's climate,down to quarks, and out to colliding black holes. Predictive simulation lies at the heart ofpolicy choices in energy and environment affecting billions of lives and expenditures oftrillions of dollars. It is also at the heart of scientific debates on the nature of matter andthe origin of the universe. The petascale is barely adequate for such demands and we arebarely established at the levels of resolution and throughput that this new scale ofcomputation affords. However, no scientific agenda worldwide is pushing the petascalefrontier on all its fronts as vigorously as SciDAC.
The breadth of this conference archive reflects the philosophy of the SciDACprogram, which was introduced as a collaboration of all of the program offices in theOffice of Science of the U.S. Department of Energy (DOE) in Fall 2001 and was renewedfor a second period of five years in Fall 2006, with additional support in certain areasfrom the DOE's National Nuclear Security Administration (NNSA) and the U.S. NationalScience Foundation (NSF). All of the projects in the SciDAC portfolio were representedat the conference and most are captured in this volume. In addition, the OrganizingCommittee incorporated into the technical program a number of computational sciencehighlights from outside of SciDAC, and, indeed, from outside of the United States.
As implied by the title, scientific discovery is the driving deliverable of the SciDACprogram, spanning the full range of the DOE Office of Science: accelerator design,astrophysics, chemistry and materials science, climate science, combustion, life science,nuclear physics, plasma physics, and subsurface physics. As articulated in theeponymous report that launched SciDAC, the computational challenges of these diverseareas are remarkably common. Each is profoundly multiscale in space and time andtherefore continues to benefit at any margin from access to the largest and fastestcomputers available. Optimality of representation and execution requires adaptive,scalable mathematical algorithms in both continuous (geometrically complex domain)and discrete (mesh and graph) aspects. Programmability and performance optimalityrequire software environments that both manage the intricate details of the underlyinghardware and abstract them for scientific users. Running effectively on remotespecialized hardware requires transparent workflow systems. Comprehending thepetascale data sets generated in such simulations requires automated tools for dataexploration and visualization. Archiving and sharing access to this data within theinevitably distributed community of leading scientists requires networked collaborativeenvironments. Each of these elements is a research and development project in its ownright. SciDAC does not replace theoretical programs oriented towards long-term basicresearch, but harvests them for contemporary, complementary state-of-the-artcomputational campaigns. By clustering researchers from applications and enablingtechnologies into coordinated, mission-driven projects, SciDAC accomplishes two endswith remarkable effectiveness: (1) it enriches the scientific perspective of bothapplications and enabling communities through mutual interaction and (2) it leveragesbetween applications solutions and effort encapsulated in software.
Though SciDAC is unique, its objective of multiscale science at extreme computationalscale is shared and approached through different programmatic mechanisms, notablyNNSA's ASC program, NSF's Cyberinfrastructure program, and DoD's CREATEprogram in the U.S., and RIKEN's computational simulation programs in Japan.Representatives of each of these programs were given the podium at SciDAC 2007 andcommunication occurred that will be valuable towards the ends of complementarity,leverage, and promulgation of best practices. The 2007 conference was graced withadditional welcome program announcements. Michael Strayer announced a new programof postdoctoral research fellowships in the enabling technologies. (The computer sciencepost-docs will be named after the late Professor Ken Kennedy, who briefly led theSciDAC project Center for Scalable Application Development Software (CScADS) untilhis untimely death in February 2007.) IBM announced its petascale BlueGene/P systemon June 26. Meanwhile, at ISC07 in Dresden, the semi-annual posting of a revised Top500 list on June 27 showed several new Top 10 systems accessible to various SciDACparticipants.
While SciDAC is dominated in 2007 by the classical scientific pursuit of understandingthrough reduction to components and isolation of causes and effects, simulation at scaleis beginning to offer something even more tantalizing: synthesis and integration ofmultiple interacting phenomena in complex systems. Indeed, the design-orientedelements of SciDAC, such as accelerator and tokamak modeling, area alreadyemphasizing multiphysics coupling, and climate science has been doing so for years inthe coupling of models of the ocean, atmosphere, ice, and land. In one of the panels atSciDAC 2007, leaders of a three-stage `progressive workshop' on exascale simulationfor energy and environment (E3), considered prospects for whole-system modeling in avariety of scientific areas within the domain of DOE related to energy, environmental,and global security. Computer vendors were invited to comment on the prospects fordelivering exascale computing systems in another panel. The daunting nature of thischallenge is summarized with the observation that the peak processing power of theentire Top 500 list of June 2007 is only 0.0052 exaflop/s. It takes the combined power ofmost of the computers on the internet today worldwide to reach 1 exaflop/s or 1018floating point operations per second.
The program of SciDAC 2007 followed a template honed by its predecessor meetings inSan Francisco in 2005 and Denver in 2006. The Boston venue permitted outreach to anumber of universities in the immediate region and throughout southern New England,including SciDAC campuses of Boston University, Harvard, and MIT, and a dozen othersincluding most of the Ivy League. Altogether 55 universities, 20 laboratories, 14 privatecompanies, 5 agencies, and 4 countries were represented among the conference andtutorial workshop participants. Approximately 47% of the conference participants werefrom government laboratories, 37% from universities, 9% from federal program offices,and 7% from industry.
Keys to the success of SciDAC 2007 were the informal poster receptions, coffee breaks,working breakfasts and lunches, and even the `Right-brain Night' featuring artisticstatements, both reverent and irreverent, by computational scientists, inspired by theirwork. The organizers thank the sponsors for their generosity in attracting participants tothese informal occasions with sumptuous snacks and beverages: AMD, Cray, DataDirect,IBM, SGI, SiCortex, and the Institute of Physics.
A conference as logistically complex as SciDAC 2007 cannot possibly and should not beexecuted primarily by the scientists, themselves. It is a great pleasure to acknowledge themany talented staff that contributed to a productive time for all participants and nearperfectadherence to schedule. Chief among them is Betsy Riley, currently detailed fromORNL to the program office in Germantown, with degrees in mathematics and computerscience, but a passion for organizing interdisciplinary scientific programs. Betsy staffedthe organizing committee during the year of telecon meetings leading up to theconference and masterminded sponsorship, invitations, and the compilation of theproceedings. Assisting her from ORNL in managing the program were Daniel Pack,Angela Beach, and Angela Fincher. Cynthia Latham of ORNL performed admirably inwebsite and graphic design for all aspects of the online and printed materials of themeeting. John Bui, John Smith, and Missy Smith of ORNL ran their customary tight shipwith respect to audio-visual execution and capture, assisted by Eric Ecklund and KeithQuinn of the Westin. Pamelia Nixon-Hartje of Ambassador Services was personallyinvaluable in getting the most out of the hotel and its staff. We thank Jeff Nichols ofORNL for managing the primary subcontract for the meeting.
The SciDAC tutorial program was a joint effort of Professor John Negele of MIT, DavidSkinner, PI of the SciDAC Outreach Center, and the SciDAC 2007 Chair. Sponsorshipfrom the Outreach Center in the form of travel scholarships for students, and of the local areaSciDAC university delegation of BU, Harvard, and MIT for food and facilities isgratefully acknowledged.
Of course, the archival success of a scientific meeting rests with the willingness of thepresenters to make the extra effort to package their field-leading science in a formsuitable for interaction with colleagues from other disciplines rather than fellowspecialists. This goal, oft-stated in the run up to the meeting, was achieved to anadmirable degree, both in the live presentations and in these proceedings. This effort isits own reward, since it leads to enhanced communication and accelerated scientificprogress.
Our greatest thanks are reserved for Michael Strayer, Associate Director for OASCR andthe Director of SciDAC, for envisioning this celebratory meeting three years ago, andsustaining it with his own enthusiasm, in order to provide a highly visible manifestationof the fruits of SciDAC. He and the other Office of Science program managers inattendance and working in Washington, DC to communicate the opportunities affordedby SciDAC deserve the gratitude of a new virtual scientific village created and cementedunder the vision of scientific discovery through advanced computing.
David E Keyes Fu Foundation Professor of Applied Mathematics
