INSTITUTION NAME: Florida State University WORK PACKAGE TITLE: IC PROJECT TITLE: Towards a Computational Science and Information Technology Curriculum to Support HPCMP Continuing Education and MSI Pipeline Development POC NAME: Geoffrey C. Fox EMAIL: gcf@cs.fsu.edu PHONE: 850 644 4587 FAX: 850 644 0098 CTA or PEI: I/C PROJECT DESCRIPTION: Educational programs in computational science are an important adjunct to the DoD's HPC Modernization Program. For many current DoD scientists and engineers, the opportunity to obtain advanced degrees provides a path for both professional and personal growth. At the same time, universities are training future generations of DoD researchers, feeding the "employment pipeline." The academic field of "computational science" is particularly relevant to the HPCMP, because it emphasizes an understanding of the computational tools and infrastructure which stand behind _all_ of the Program's ten CTAs. Working toward an advanced degree in computational science provides practicing S&Es a route to deepen their knowledge of the computational techniques undelying their chosen field of research. Likewise undergraduate programs in computational science help to prepare students for the kinds of problems and tools they will encounter if they pursue a career in a computational science field. This kind of preparation is rarely found in current Computer Science curricula, nor in those of the traditional science and engineering departments. Further we learnt from experiences at Syracuse that it is very helpful to generalize classic computational science (which is aimed at scientific simulations) to include information technology. The newly established School of Computational Science and Information Technology (CSIT) at Florida State University (FSU) will soon begin one of the few Computational Science degree programs in the country. Classes are expected to start in Fall 2000 (for students in other disciplines), and students will be allowed to enroll in the CSIT degree program in Fall 2001. Because of this new program, and the Unversity's general commitment to the use of distance education, there is a unique opportunity to bring advanced degrees in Computational Science closer to interested HPCMP researchers. By developing courseware for distance delivery, it would become possible for DoD S&Es to take most of the required coursework towards an advanced degree while continuing to work. In some cases, even portions of the dissertation research would be related to work activities and could be conducted at the work site rather than in residence at the University. In this way, it would be possible to obtain an advanced research degree while minimizing the the amount of leave that must be taken from work. This courseware could also be adapted for incorporation into undergraduate curricula for use at other institutions. The planned CSIT curriculum is particularly well-suited for use with both DoD S&Es and Minority Serving Institutions, as it stresses practical, fundamental computer science together with a broad range of applications. It includes both classic high performance computing and the information technology curricula which has broad appeal to MSI students and for HPCMP allows one to address the full range of CTA's including simulation dominated areas like CFD and information based areas like FMS and IMT. It is essentially the same curriculum which is known as "Internetics" at Syracuse and Peking Universities. The CSIT program will offer both Masters and PhD degrees where students can choose a simulation or information emphasis, but all students will know the basics of both tracks. The core requirement for these degrees will be three of the following four courses: Computational Science I This is the first course in a two-semester sequence on the role of computational methods, models of computation, computer architectures, and digital computations in scientific applications. Cross-cutting scientific applications selected from topics in materials science, quantum mechanics, data analysis, global warming, earth-quake propagation, will be used to underscore the importance of numerical methods in linear algebra, ordinary and partial differential equations, Monte Carlo methods and tools for software development, performance monitoring, visualization and model evaluation. Computational Science II This is the second course in a two-semester sequence on the role of computational methods, models of computation, computer architectures, and digital computations in scientific applications. Cross-cutting scientific applications selected from topics in fluid flow, phase transistions, biological structure and climate prediction will be used to underscore the importance of parallel computing, numerical methods in linear algebra, ordinary and partial differential equations, Monte Carlo methods, and optimization techniques. Applied Information Technology I This is the first course in a two-semester sequence covering the application of information technologies of current interest within integrated online environments for distributed scientific computing, online research collaborations, education and electronic commerce. Information technology applications currently in the course include collaborative research environments, genomics database application, electronic logbooks, weather prediction, and teaching and learning systems, but specific applications will evolve rapidly to insure inclusion of leading-edge scientific applications of information technology. Applied Information Technology II This is the second semester of a two-semester sequence covering the application of information technologies of current interest within integrated online environments for distributed scientific computing, online research collaborations, education and electronic commerce. Information technology applications will be selected from distributed physics simulations, remote instrument operation and control and advanced teaching and learning systems, but specific applications will evolve rapidly to insure inclusion of leading-edge scientific applications of information technology. Our plan for Year 5 is to investigate the interest among both MSIs and DoD S&Es and decide on a detaled implementation plan, using distance education combined with thesis work at FSU. Visiting instructors/mentors at DoD sites or MSI from FSU may also be possible. If we think of this within the "Thompson Model for PET", onsite personnel from university partners could be important here. Simultaneously, we plan to develop asynchronous distance-delivery versions of one of the core courses described above for initial evaluation (we expect to use RealNetworks technologies here). In conjunction with this work, we will also assess new standards in the area of training materials (IMS from Educause and SCORM from DoD's Advanced Distributed Learning program) to understand how they effect PET training and education activities. We will develop guidance regarding producing training materials compliant with these standards, and by way of example, we will make selected modules of our courseware compliant. PROJECT OBJECTIVES: In conjunction with the implementation of a new graduate degree programs in the School of Computational Science and Information technology at Florida State University (CSIT), we plan to migrate the CSIT courseware to Web and CD-ROM form that will allow this program to be easily extended to include DoD S&E's by minimizing the amount of time that would have to be spend on campus at FSU to complete a graduate degree. The same courseware can be adapted to the needs of minority-serving institutions and used to enhance their curricula used to train future HPC researchers. DELIVERABLES: 1) A whitepaper laying out an implemenentation plan designed to make CSIT's degree programs accessible to DoD researchers and useable by MSIs. 2) Distance-deliverable versions of one of the FSU CSIT core courses 3) A whitepaper analyzing the relationship of IMS, SCORM, and other training standards to PET activities CUSTOMERS/END USERS: DoD S&Es interested in obtaining a graduate degree in Computational Science and Information Technology. MSIs using the courseware to enhance their curricula and improve the training of future researchers. BENEFIT TO THE WARFIGHTER: The need for extended periods of residence on a university campus is one of the biggest impediments to established professionals pursuing higher degrees. This project will make it easier for DoD S&Es to pursue advanced degrees in computational science. PROJECT DEPENDENCIES AND SCOPE: This project can easily be coupled to an effort to adapt the material for use at MSIs. Such a project would be beneficial, in that the MSI participants could not only provide feedback about courseware, but also help contribute to its development. However this project does not depend on being linked to an MSI effort. It is important to note that the formal FSU degree program is not expected to accept students until Fall 2001 (during Year 6). In Year 5 project focuses on adapting FSU courseware for HPCMP/MSI distance delivery, and on preparing the target DoD community for the idea. RISK ELEMENT: The biggest risk lies in persuading the DoD community that it is both possible and desirable to pursue an advanced degree in Computational Science and Information Technology from FSU with a minimum on-campus residence time. we have confirmed with FSU that the university is very interested in this program. We expect that a small number of interested candidates will appear initially, but once they (and the program) demonstrate success, more will follow. REQUIRED FUNDING LEVEL Year X: Year X+1: Year X+2: --------------------------------------------------------------------- -- David E. Bernholdt | Email: bernhold@npac.syr.edu Northeast Parallel Architectures Center | Phone: +1 315 443 3857 111 College Place, Syracuse University | Fax: +1 315 443 1973 Syracuse, NY 13244-4100 | URL: http://www.npac.syr.edu