Focused Effort Title: HPCC Educational CD-ROM Thematic Area(s): HPC Training and DoD User Productivity PI Name: David Bernholdt PI EMail Address: bernhold@npac.syr.edu PI Telephone: 315 443 3857 PI Fax: 315 443 1973 Project Description: NPAC already has the core of such a collection in a CD-ROM that was produced for our Computational Science Education work in China (http://www.npac.syr.edu/projects/pcrc/cpswt-e/). We would remove web technologies (except for Java for computational science) and include Virtual Programming Laboratory and SciVis material. We would then augment this with additional materials (i.e. grids, algebra, programming tools) drawn from other PET partners, guided by an editorial board including (we suggest) Jack Dongarra, Geoffrey Fox, Chuck Koelbel, Joe Thompson, Louis Turcotte. We envision this as an on-going project, with regular releases new CDs containing updated materials in order to keep abreast of evolving tools and technologies. The current proposal includes funds for NPAC to update materials, coordinate the acquisition of additional materials, and produce the first CD distribution. It also includes funds to support other contributors updating and formatting their materials appropriately for this method of dissemination. As a practical matter, we expect these funds to go directly from NRC to the appropriate universities, thus avoiding additional costs and delays. Benefits: This projects is intended to provide a broad range of HPCC educational material to DoD researchers to augment the more specific training classes available at the MSRCs. It will help researchers make better use of the HPCC resources, and help them better understand which trainings can be most helpful to them. Deliverables: CD-ROM produced and ready for distribution (March 1998) Customer Commitment/Involvement: This project was specifically requested by Louis Turcotte. ---------- Focused Effort Title: Parallel CMS Planning Thematic Area(s): Scalable Computing Migration PI Name: David Bernholdt PI EMail Address: bernhold@npac.syr.edu PI Telephone: 315 443 3857 PI Fax: 315 443 1973 Project Description: The Comprehensive Mine Simulator (CMS) is an important FMS application which currently runs on (uniprocessor) SGI systems. Currently the system can handle 30,000-50,000 mines on a single host, but has a target of 1,000,000 mines which clearly cannot be achieved without migrating to HPC systems. This project will enable us a chance to examine the application and work with the developers in order to understand the detailed requirements and effort required to parallelize this application. This would allow us to reliably scope and plan for the actual parallelization, which would be a follow-on project. We will work with Steve Bishop's group at the Night Vision Directorate, Ft. Belvoir, VA to define a reasonable collaborative approach to the actual parallelization effort. Benefits: This project was recommended by the FMS CTA lead to increase the usability of HPC systems by FMS researchers. Deliverables: Report and Project Proposal for CMS parallelization effort (March 1998) Customer Commitment/Involvement: Based on information provided by the FMS CTA lead, we plan to work Steve Bishop, Night Vision Directorate, Ft. Belvoir, VA and anticipate that the follow-on parallelization project will be a joint effort. ---------- Focused Effort Title: Object Web RTI Prototype Thematic Area(s): Scalable Computing Migration HPC Training and DoD User Productivity HPC Performance Metrics/Tools Management and Interpretation of Large Data Sets SciVis for Very Large Problems DoD Challenge Applications PI Name: David Bernholdt PI EMail Address: bernhold@npac.syr.edu PI Telephone: 315 443 3857 PI Fax: 315 443 1973 Project Description: DMSO introduced recently new integration framework for advanced simulation networking called High Level Architecture (HLA) and based on Run-Time Infrastructure (RTI) software bus model. RTI enables federations of real-time/time-stepped and logical-time/event-driven simulations/federates and it optimizes communication via event filtering and publisher/subscriber region/interest matching, supported by the Data Distribution Management (DDM) service. Full and rapid DoD-wide transition to the HLA is strongly advocated by DMSO and facilitated by open public specifications of all HLA components, extensive nation-wide tutorial programs and prototype RTI implementations. In particular, early prototype codes for RTI F.0 ("familiarization" release) and RTI 1.0 subset were released by DMSO to the DoD simulation community in spring/fall '97. The complete RTI 1.0 implementation, including full DDM support is expected by the end of '97. Given the systematic shift of the DoD training, testing and wargaming activities from the physical to synthetic environments, and the ever increasing computational demands imposed on advanced modeling and simulation systems, high performance distributed computing support for HLA will likely play the crucial role in the DoD Modernization Program. Based on our expertise at NPAC in Web/Commodity Software based HPCC, we propose a focused project within the HPCMP/PET program that will adress a critial aspect of rapid deployment of high performance HLA systems: World-wide distributed RTI on top of Java/CORBA based Object Web and focused on Web based collaborative HLA training The DIS->HLA transition process is rapid and, for some people often complex or/and confusing. DMSO provides significant help to the community in terms of high level tutorials and prototype RTI codes. Additional help, as proposed in this project, could be provided by an open, source level RTI subset implementation using Object Web/Commodity technologies such as CORBA and Java. Such RTI would be made broadly accessible both at the source code and runtime/collaborative simulation level, and integrated into the existing and forthcoming HLA training and tutorial suites. At NPAC, we are currently developing Java based Web Object Request Broker (WORB) server that will support HTTP and IIOP protocols and will act as a univeral node of our HPcc (High Performance commodity computing) environment. Given that RTI object bus model is strongly influenced by CORBA and DMSO is in fact interacting with OMG towards proposing HLA as CORBA simulation facility/framework, an early Java/CORBA based RTI prototype seems to be a natural effort in the domain of interactive HLA training. Our Object Web/WORB based RTI subset would support and integrate Web DIS (Java and VRML based) applications under development at the Naval Postgradual School at Monterey, CA, as well as more traditional and substantial simulation codes such as ModSAF and perhaps also SPEEDES, TEMPO or IMPORT, currently at the planning stage as possible FMS training targets for our PET activities at ARL. By the end of this project, we will deliver a prototype object web (CORBA) based RTI kernel (subset), capable of running a a simple demonstration application to be developed locally. This would serve as a demonstration of the integration of DMSO and web technologies and provide a freely available tool. A follow-on project could further develop the system into a full FTI implementation, at which point it would be possible to run real RTI applications as a demonstration of this tool. Benefits: This project will enhance the ability of the FMS community to make use of HPCC resources. Deliverables: CORBA/HPCC based implementation of RTI kernel (subset) with simple application for demonstration purposes (March 1998) Customer Commitment/Involvement: The FMS CTA lead has requested parallelization of RTI and examination of web technologies in this context. ---------- Focused Effort Proposal for PET Title: Distance Education for Curricula Enhancement at HBCUs PI Name: David Bernholdt PI EMail: bernhold@npac.syr.edu PI Telephone: 315 443 3857 PI Fax: 315 443 1973 Vision Distance education is likely to be revolutionized by the growing power of high speed networks and Web Software for delivery, preparation and teacher-student/student-student collaboration components of education and training. This impact will be important at all levels of education K-12, undergraduate, graduate, continuing education and institutional training as in the DoD. Currently there is very little experience as to what features of both hardware and software are important and it is probable that there is no one answer but different applications will have different requirements and varying success. NPAC has in TANGO and WebWisdom leading edge collaboration and dissemination technologies. Further the Virtual Programming Laboratory is a good prototype Web lab environment. Further we have substantial experience with computer Science graduate and undergraduate classes, Physics 105/106 undergraduate class, Living SchoolBook at K-12 level and Cornell virtual workshop collaboration for training in the use of the Web for education and training in many areas. Project Description and Value to PET The application of NPAC technology can impact both asynchronous (self study by student) and synchronous (teacher delivery) modes. Further it can be used both in MSRC training and education as well as in programs with the partnering HBCU's. The MSRC applications could be a model for much broader scale DoD training and so lessons learnt could be of profound importance. Again the HBCU applications are a model for how knowledge(from "leading edge research" Universities) in rapidly changing fields can be transferred to the curricula of educational institutions which are not so strongly coupled to research in these fields and so find it hard to keep their curricula up to date. Thus impact includes education in community colleges and similar institutions and will have broad national importance and not just confined to a few or all HBCU's. This proposal covers the initial phases of what is anticipated to be a long-term Focused Effort in Distance Education. The goal at this stage is to setup the necessary infrastructure to provide distance education at CEWES and the HBCUs and to conduct several trials designed to help evaluate different strategic approaches and technical details of the implementations. It is worth noting that taken individually, not all of these trials will be considered "successes" by those involved. The real measure of success must be developing from these trials an effective set of strategies and implementations for distance education which can be applied as this Focused Effort continues. Here the goal is not just to educate a few students better but to leave a lasting impact on the institution in terms of a better curricula and more knowledgeable faculty equipped with better Web based tools. This implies project must have committed involvement of HBCU faculty. We also rely on the established relationship of CEWES & PET management with the HBCUs. Consequently, we offer not a list of detailed deliverables, but an overall plan, with consultation among all of the partners to fix the details as the project proceeds. In light of the fluidity required by this project, the staffing and budget estimates are no more than rough estimates. As presented, these figures suppose strong involvement with both Jackson State University and Clark Atlanta University. They also rely on the reuse of course materials already developed by NPAC for these initial experiments--new courses can certainly be developed, but this will require adjustments to schedules and budgets. These assumptions are certainly subject to discussion and revision as the details of the plan are fixed. Deliverables NPAC will take primary responsibility for these items, but specific deliverables must be designed and agreed by all partners. * Ongoing enhancement of Tango, WebWisdom (and if involved the Virtual Workshop) to support the selected distance learning experiments. * A compact (1-2 week) presentation of a course involving only Syracuse and CEWES as a "shakedown" run for the delivery systems. Course topic will be decided in consultation with CEWES management, but a basic web technology course is anticipated. (by August 1997) * Once technical glitches are worked out, one more compact (1-2 week) course should be presented, possibly at one of the HBCUs rather than at CEWES. (by December 1997) * Once technical details are ironed out, Syracuse will coordinate with CEWES management and HBCU faculty to plan a set of distance learning experiments to be carried out during the remainder of the year and continuing into subsequent years. (by December 1997) * Once HBCU implementation starts, we suggest several different approaches could be tried with perhaps the different universities choosing to try out different learning models. Important issues to explore include: o Tradeoffs between synchronous and asynchronous delivery o Role of professors at "client" site (i.e. HBCU) could vary from traditional synchronous deliverer to being a mentor for students with remote University (e.g. NPAC) delivering material. o "Train the trainers" (i.e. professors at HBCU) versus "Train the Students" implementation. o Importance of the exploratory aspect of learning which is naturally supported by Web-based approach. * Regular semester-length courses to be offered at HBCUs starting January 1998. * While the ultimate goal of distance education is to minimize the need for face-to-face student-instructor contact, the nature of this project is such that we consider it advisable to make somewhat more on-site visits than we would expect in the long term. For the compact courses, we plan 1 day at the beginning of the course and one day at the end for followup. For the semester-length courses, we plan for trips of 1-2 day duration at the beginning, middle, and end of the course. Additional Requirements These items are considered critical to the success of this project, but while NPAC hopes to play a significant role in the decision-making process, they would seem ultimately to be the primary responsibility of our partners. * Installation of appropriate networked multimedia PC's at HBCU's in full classroom configurations (at least 24(12) machines with one(two) students per node). Smaller clusters with identical base hardware should be present at CEWES, NPAC and other relevant places. These systems should be linked by high speed networks supporting interactive video conferencing as well as multicast of educational videos served from a digital video server. * Hiring of a "Network Training Lead" by Jackson State University ---------- Title: Interfacing Databases and the Web: Management of Large WWW Sites Using Commercial Database Technology PI Name: David Bernholdt PI EMail: bernhold@npac.syr.edu PI Telephone: 315 443 3857 PI Fax: 315 443 1973 Project Description and Value to PET This project couples commercial relational database (RDBMS) technology with the world-wide web to make it easier to effectively the large amounts of diverse information which may be presented on the WWW site of a large organization such as the CEWES MSRC. There are three basic components of the project. 1. Integration of existing databases with the WWW in order to be able to present data obtained by querying the database, insuring that the WWW site always reflects the most current information available in the database. 2. Use of an RDBMS to "back-end" the WWW site, holding all web pages which do not originate from the aforementioned databases or other specialized systems, as well as "metadata" about the web pages, including creator, date of last update, expiration date of information, review status for public release, etc. 3. A suite of tools to facilitate management of the WWW site, taking advantage of the RDBMS and the metadata it holds. Capabilities will include summary information about the structure and size of the site, identification of "dead" hyperlinks and "orphaned" web pages as well as outdated information. The system will support multiple levels of access control and the ability to "stage" information into the production part of the service to accomodate review procedures and ease of development. Detailed design requirements, following this basic model will be developed in cooperation with designated CEWES staff at the beginning of this project. Because we anticipate that the basic techniques of interfacing commercial database systems to the WWW can be used elsewhere in CEWES and the PET program, we also offer a training course on this topic. Deliverables * A detailed design for the WWW management system will be developed in cooperation with designated CEWES staff (six weeks after commencement of project) * A prototype of the management system will be developed and installed at CEWES for evaluation and feedback (by six months) * Final version of the WWW management system will be installed at CEWES (by nine months) * A 1-2 day training course on parallel and web-linked databases (prepared for delivery after August 1997) ---------- Title: Web-Linked Databases for Domain-Specific Information Repositories PI Name: David Bernholdt PI EMail: bernhold@npac.syr.edu PI Telephone: 315 443 3857 PI Fax: 315 443 1973 Vision The ever-increasing volume of information available electronically, especially via the Internet, can be of great value to DoD researchers, but it is becoming increasingly easy to be inundated by all of the information. Increased coupling the popular dissemination technologies (i.e. the World-Wide Web) with tools designed to deal with large amounts of information (i.e. commercial database management systems) will be not only useful, but eventually invaluable in helping DoD researchers deal effectively with the glut of information. Project Description Domain-specific search engines represent a scalable alternative to attempting to maintain a global search capability for the World-Wide Web (which the popular WWW search engines are finding increasingly hard to do). With the help of domain specialists to identify appropriate targets for indexing, a domain-specific search engine can provide DoD researchers with information repositories which are more easily accessible, can be kept more current, and are more efficient to use than global search engines. These information repositories can stand alone, or can compliment other resources such as software repositories, database-backed archives of topical mailing lists, etc. This project will deploy the infrastructure to create domain-specific WWW search engines at CEWES and assist in the development of search engines in two domains to demonstrate the power of this idea. The subject of grid generation is of great importance to the CFD and CSM CTAs as well as some CWO models, areas which consume the majority of the CEWES MSRC's computational resources. The grid generation search engine will compliment the CTA-specific software repositories which are being developed under a cross-MSRC initiative, also lead by NPAC. The other area, which is of great importance to the Corps of Engineers, is geographical information systems (GIS). In this case, the repository will stand alone, but we envision that it may form the core of a more sophisticated information repository in the future, perhaps including a GIS software repository and other database-backed tools. Deliverables * Installation and setup of NPAC's database-backed WWW indexing engine on a designated machine at CEWES. (Oracle RDBMS installation to be provided by CEWES) (by 2 months after project start) * Development and operation of a WWW index on grid generation technologies, based on input from domain experts (Joe Thompson, and others) (by 4 months after project start) * Development and operation of a WWW index on geographical information systems (GIS), based on input provided by domain experts (to be designated by CEWES) (by 6 months after project start) ---------- Focused Effort Title: Tango for Collaboration and Distance Consulting Thematic Area(s): HPC Training and DoD User Productivity PI Name: David Bernholdt PI EMail Address: bernhold@npac.syr.edu PI Telephone: 315 443 3857 PI Fax: 315 443 1973 Project Description: The Tango collaboration system has been used successfully in another CEWES Focused Effort to support collaboration in the form of remote classroom instruction, demonstrating the efficacy of such a system for use within the DoD. This proposal seeks to further extend the application of the Tango collaborative framework to another area of great importance to the DoD and the PET program: technical collaboration among geographically distributed groups, especially in the area of HPCC software development and consulting (i.e. PET or MSRC staff working with remote HPCC users). Tango already includes basic collaborative tools, such as audio/video conferencing, shared whiteboard, chat tool, etc. This project will augment the system with tools necessary to support software development, performance improvement, and related activities. Early work will include detailed training in the operation and maintenance of the Tango server for an appropriate CEWES staffer and installation of the basic Tango system (including server) at CEWES making it available to MSRC users for basic collaboration. Development of the additional tools will be done in cooperation with an appropriate group (to be identified), and at the end of the project, the entire system will be available for general use. Benefits: This project represents a further step in trying to reduce the "importance of place" in access to MSRC resources and MSRC and PET services, as well as in DoD research efforts as a whole. Deliverables: * Tango server will be deployed on a suitable host at CEWES * Training in the operation and maintenance of the Tango server will be provided at Syracuse University * A training at CEWES to familiarize users with the operation and use of the Tango system. * Enhancements to Tango suitable for collaborative software development and consulting, including * Shared source code viewer/modifier: we will develop a collaborative version of emacs, a public domain editor with very strong support for code developers.We will provide extensions for collaborative viewing/editing and extend support for a versioning to synchronous collaborative mode. Emacs support for the basic development cycle (code - compile - execute - debug) will be extended to the collaborative mode so that all these steps can be shared by all team members. * Shared flow diagramming tool: we will extend functionality of our shared whiteboards to fully object-oriented, editable, shared drawing tool. This tool will enable users to work on graphical representations of their computational processes, such as flow diagrams and entity relationships models. * Shared debugger: we will implement a tool allowing for distribution of the results of a code debugger, either real-time or post-mortem. * Shared numerical results visualizer: we will provide the infrastructure supporting shared, distributed, real-time visualization of the numerical results produced by codes running on HPCC platforms. The framework will consist of a runtime for transparent, distributed visualization and of the application programmer interface allowing for an easy hookup of the codes to the visualization system. The runtime will provide a reach set of numerical filters and the store-and-animate capability for analyzing code numerical stability or for tracing numerical convergence processes. A prototype of such a tool has been already implemented in NPAC. Written entirely in Java, the framework is extremely portable and easy to use. * Java MPI performance visualization: we will study the feasibility of building a sharable, distributed interface to the tools providing basic performance metrics of HPCC platforms. We will select the most promising performance visulaization packages from among public domain systems such as AIMS from NASA or Pablo from University of Illinois, or commercial systems such as Datametrix, IBM perfmon, or SGI Co-Pilot. Customer Commitment/Involvement: CEWES has already expressed a desire for a full-scale Tango installation (including server), and for training in the care and feeding of such a system. This project also requires a group willing to help with the design and testing of the new remote consulting/software development tools for Tango, but there are many candidates and we do not anticiapte problems identifying a suitable candidate.