Sep 18 2008
A Definition and Perspective on Computational Sciences
Just a snippet from a grant we are submitting for a new cluster computing instrument for TAMUC, that I was a bit proud of writing. Perhaps undergraduate and graduate students might get a better feel for some of the current and near term work you might be involved with me and the sciences at TAMUC from this description.
We have historically divided the scientific endeavor into three fundamental domains: the physical sciences, focusing on nonliving matter (physics, chemistry); the life sciences, focusing on living matter (biology, genetics); and the social sciences, focusing on humans and their societies (psychology, cognitive science). With the creation of computers, it can be argued that a fourth fundamental domain of science has been created that focuses on computation (Rosenbloom, 2004). Whether or not one agrees that computation is a new fourth fundamental domain of science, computational modeling and scientific computing has added a new fundamental tool to the scientific toolkit, augmenting the traditional tools of analytical methods (mathematical descriptions), and experimentation. High performance scientific computing allows for large-scale computational models, which fit a point midway between analytical methods and experimentation. Large-scale computational models allow scientists to explore problems much to complex to be described purely analytically, for real world complex dynamical systems. But they also allow for a type of virtual experimentation of sufficiently complex problems, but without the danger or expense that might be needed to conduct such experiments in the real world.
The intersection of computer modeling with the traditional sciences has opened up a whole new research methodology. Increasingly, it is the intersection between traditional sciences, along with computational modeling, that are driving important scientific discoveries. For example, computation along with biology and chemistry drive much of molecular biology and genetic sequencing research, providing fundamental insights into disease processes and health issues. In our proposal, you will see many other examples of these types of synergies being performed at Texas A&M University – Commerce (TAMUC), and how computational methodologies are crucial to their success. Because of these increasingly important synergies, there is a great demand for trained and knowledgeable professionals with backgrounds both in a fundamental scientific domain, as well as experience with high performance computing and computational modeling. At TAMUC we are participating in and helping to drive this paradigm shift, and are heavily involved in the training, education and implementation of these computational methods for attacking fundamental scientific questions. As with many such organizations, our research efforts started out by building computational infrastructures in isolation. We have realized, however, that a critical mass now exists on our campus of research in this area, as well as an increasing demand from industry and technology firms in our region for trained researchers in these methods. The acquisition of a high-performance cluster computing infrastructure will serve as a focal point for uniting these efforts on campus and in our region, and developing a focused training and research group agenda in the computational sciences.