LLNL Home S&TR Home Subscribe to S&TR Send Us Your Comments S&TR Index
Spacer Gif

Spacer Gif

William H. Goldstein

Wayne Shotts
Associate Director of Nonproliferation, Arms Control, and
International Security and Acting Director of the Homeland Security Organization

Computing Science: One Arrow in the Quiver for Homeland Security

WHEN the legislation establishing the Department of Homeland Security was being drafted, eyebrows were raised across the country over the explicit inclusion of Lawrence Livermore’s Advanced Scientific Computing Research Program. The program was transferred from the Department of Energy’s Office of Science to the new department’s Science and Technology directorate.
With so much attention being directed at top-level issues such as transportation security and biodefense, the role of scientific computing in homeland security is often overlooked. All of the major homeland security issues have critical elements that require advanced computing. Advanced scientific computing is more than the world’s fastest or largest computers, the aspect of computing for which Livermore is typically known.
In many cases, the homeland security computing challenges involve new modeling techniques, new algorithms for handling data, or even new ways of thinking about a problem. Once these techniques are in hand, they can often be implemented on ordinary desktop computers or palm-sized instruments, making it possible to deliver state-of-the-art computing capabilities into the field and and even to first responders.
The advanced computing challenge most in the news is the intelligence problem of finding and connecting the critical pieces of information amidst mountains of data. Advanced scientific computing is key to developing the scalable information analysis techniques and large-scale data integration tools that are needed to solve this problem. In addition to developing new algorithms, we are also exploring novel computing architectures that could dramatically speed the discovery of relationships in large data sets.
The detailed simulations and scenarios required to improve emergency response planning and crisis management operations depend on advanced computing capabilities. Lawrence Livermore has long been a leader in computational fluid dynamics and other simulation technologies. This work forms the foundation of the National Atmospheric Research Advisory Center (NARAC) and is being applied to extend NARAC’s capabilities to model the fate and transport of chemical and biological agents in the atmosphere. For example, we are applying expertise in adaptive mesh refinement and complex geometry modeling to simulations of aerosol dispersion in urban settings.
Advanced computing is also required for the comprehensive and dynamic risk-vulnerability-mitigation assessments and consequence analyses that are needed to develop strategies for protecting critical U.S. infrastructures. For instance, by combining the Laboratory’s simulation strengths in shock physics and structural mechanics, we can understand the impact of explosions on buildings and other structures and then design appropriate countermeasures.
Advanced computing contributes to the development of better radiation detectors and biosensors. If we are to design detectors that exploit signatures that terrorists think are undetectable, we must first be able to model the physical and biological phenomena involved. In some cases, this requires first-principles simulations on supercomputers. Advanced computations also guide the micro- and nanoscale engineering required to miniaturize the devices.
As the article On the Front Lines of Biodefense describes, advanced computing techniques are also being used to speed the development of the DNA and protein signatures that lie at the heart of biodefense. For example, biologists, computing scientists, and mathematicians collaborated to invent the first-ever algorithm for aligning draft genomes with finished genomes, making it possible to identify candidate signatures from incomplete data and thereby enable rapid responses to unexpected disease outbreaks. This kind of multidisciplinary teaming is a hallmark of the way Lawrence Livermore tackles tough problems. Other computing science advances have significantly reduced the time of a key step in the signature development process, and the algorithms involved are being scaled to permit work on organisms with larger genomes. Taken together, these computing advances will result in faster, less expensive, and more reliable biodetection capabilities for homeland security.
Lawrence Livermore is a leader in providing science and technology for homeland security. An important element of our success is our ability to integrate multiple scientific and technical disciplines to produce systems-level solutions to national security challenges. Advanced scientific computing is one of the keys to “being smart” about how we tackle homeland defense, enabling us to provide technologies and capabilities that significantly improve security.

Back | S&TR Home | LLNL Home | Help | Phone Book | Comments
Site designed and maintained by TID’s Internet Publishing Team

Lawrence Livermore National Laboratory
Operated by the University of California for the U.S. Department of Energy

UCRL-52000-04-4 | April 6, 2004