MICS Technical Advisor

The purpose of this assignment is to provide support for various efforts such as the Coordinating Committee for Informatics, Research, Development, and Applications (CCIRDA), the Small Business Innovative Research (SBIR) program and miscellaneous reviews.

Information Systems

The purpose of the collaboratory management environment project is to build upon expertise gained through two pilot projects at Ames Laboratory and Oak Ridge National Laboratory, to design and implement a web-based management environment. The tasks associated with combining the functionality of the pilot projects include developing generic web-based interfaces for financial databases, defining models for project databases at DOE sites, implementing flexible security and domain restriction features, developing graphical web-based interfaces that support multiple/ matrix views, designing local intelligent software agents to support autonomous information access and analysis, identifying and incorporating searching/indexing capabilities and defining query mechanisms, and integrating all features into an initial prototype system for DOE/MICS and for two laboratories: Oak Ridge National Laboratory and Ames Laboratory. This system will allow cross-cut and filtering mechanisms and capabilities that do not currently exist to integrate data from heterogeneous databases and documents distributed across the WWW into precise, meaningful formats.

Public Key Distribution

Secure communication channels will be a critical component of any distributed collaborative environment such as the Distributed Informatics, Computing and Collaborative Environment (DICCE) project. The successful management and operation of such a system requires reliable authentication and authorization tools. Public key technology is the key to solving these problems and a scalable method for the distribution of public keys has yet to be resolved. This work will develop and implement software that addresses the key distribution issue in a scalable manner.

20e. APPROACH

MICS Technical Advisor

Key Personnel - FY1996 S. Elbert (PI) (0.3 FTE)

Information Systems

The research agenda presented in this section is crucial to creating a useful and powerful tool from today's web browsing tools. Accomplishing the tasks outlined will establish the framework for a robust, scalable, and secure virtual management system that could ultimately become the de facto standard management system for the Department of Energy. This system will provide sophisticated search and cross-cut capabilities within a single site or across multiple sites for both finance and project information. For example, with this system in place, a program manager at headquarters can rapidly respond to requests from the Secretary for research highlights for an upcoming speech or from (OMB) for budget information. In response to the Secretary's request, a search can be initiated at the appropriate research sites for the latest information. Since the system will be dynamic, researchers in the field will have updated their progress with visualizations, including graphs and animations, and the corresponding text, and program managers will have rapid access to the most recent progress information. When preparing the response for OMB, past financial reports from across the various sites could be accessed to identify progress, trends, and/or areas for cost savings.

Key Personnel - FY1996 B. Helland (PI) (0.1 FTE); N. Imani (PI) (0.5 FTE)

Public Key Distribution

The Pretty Good Privacy (PGP) approach to privacy enhanced e-mail developed by Philip Zimmermann is the most widely used method of handling secure e-mail available today. Unfortunately, PGP does not provide enough information to locate a key, so a key server must provide this function. The existing server, developed by Michael Graff, has major shortcomings in terms of scalability, speed and PGP dependency. Graff has designed a second generation server that addresses these issues and others, such as X.509 use. This server will be implemented and made available to the ESnet community.

Contributing Principal Investigators - FY1996 S. Elbert

20f. TECHNICAL PROGRESS

Detail Assignment

CCIRDA meetings were held October 11, 1994, March 7, 1995, and November 6, 1995. The next meeting will be held April 2, 1996, when a change in the structure of CCIRDA, as reflected in a new charter, will be proposed.

Information Systems

During FY1996, an initial prototype, based on expertise gained in the pilot projects will be developed and tested at Ames Laboratory, Oak Ridge National Laboratory and DOE headquarters. During the design and implementation phase, work on the prototype will be divided into two functional components that will support the web-based information system: intelligent agents at each site and data analysis and programming tools.

Public Key Distribution

Implementation of the new design began in January, 1996. By the end of March, 1996, the initial implementation will be completed and distributed to NERSC, ANL, and other ESnet sites for testing. The new server should be operational by July, 1996.

20g. FUTURE ACCOMPLISHMENTS

MICS Technical Advisor

CCIRDA will become a self-formed group of Laboratory CIO's that will act as a forum for information exchange, consensus building, and coordination of the major non-federal scientific activities and issues in computing and information processing. It will provide feedback and proposed initiative direction for activities affecting the Laboratories and will partner with DOE in developing an integrated Information Technology long range plan. Furthermore, it will identify issues and key emerging technologies to support Laboratory programmatic missions and administration.

Information Systems

Research Tasks:

Study current operating process for proposal submission and project management to identify critical features

Since an interagency task force is evaluating the proposal submission process, the efforts to integrate the processes noted in this proposal are synergistic with the efforts of the interagency task force. Working with the information technology model provided by the interagency task force for the proposal submission process, the logical links into the project and financial data bases must be identified and incorporated into the integrated management environment. In addition, project and financial data bases at different sites are characterized by large, legacy systems. We will evaluate the project management processes to determine the critical path from the overview at the highest sponsor level to the detailed project management level at participating sites. This process must be understood to build an information technology model that logically integrates each of the processes.

Define functionality for data analysis and programming tools that support the web-based interface.

The web-based interface provides a mechanism for seamless access to information that a requester has the right to access. This is the mechanism by which a user makes queries via the intelligent agent for information in the financial and project data bases. The query can be for cross-cut information within a site or across sites, information that has changed since a previous query, high-level roll ups or low-level detailed information. We may find it necessary to customize the web-based interface depending on the level or affiliation of the user.

As noted earlier, currently the WWW does not provide support to search on keywords and concepts to provide cross-cuts into the data. Initially, the system being proposed must be able to search on keywords and concepts to provide cross-cut information within a single site or across selected multiple sites. Examples queries are: search for references to "distributed computing" across all sites, list all papers and technical reports for a specified Budget and Reporting code across two sites, or provide highlights across all sites since a specified date. The local intelligent agents will facilitate efficient indexing and querying capabilities. The proposed system will allow data to be rolled up to high-level cross-cuts from the distributed systems as an overview. The system will also incorporate a drill-down approach where traversing capabilities will provide access to detailed information at the different sites based upon the appropriate and authenticated access level for the user.

The web-based interface will be built via innovative use of viewers, such as Mosaic and Netscape, which should provide for the broadest spectrum of users and facilitate future maintenance of the finished system. Research concepts identified in this project can be transferred to industry to promote efficient consumption of resources.

Define meta descriptor and functions of intelligent agent

The intelligent agent at each site will provide a standard interface into the financial databases and project descriptions. This agent will communicate with each site via a meta descriptor that defines the structure of the information which contains at least the following: field work proposals, personnel, highlights, technical reports and published papers. The project information should also be defined to accommodate text, audio, and video information. One important function of the meta descriptor will be to define hyperlinks between the financial data and project information. The intelligent agent must also function as the first line of security into the system. The agent will be designed with the flexibility to allow Program Managers at DOE headquarters to access information pertaining to his or her program but not to someone else's program. Furthermore the agent should allow sites to restrict access to some information, but allow broader access to other, perhaps collaboratory, information. The meta descriptor will also identify information that requires secure transmission over the network. Other functions of the intelligent agent include automatically uploading new information to the end user system, if appropriate, flagging new items since the last query, and archiving old items that have been replaced.

The intelligent agents and collaborative financial and project tools coupled with the web-based interfaces provide the framework to logically link the financial, project, and dynamic proposal information with the static information provided during proposal submission process. Through exploiting the power of these innovative technologies, a virtual management environment that provides managers with seamless access to distributed information in an integrated, complete, and timely manner from a single source is possible. To be useful, information must be both relevant and reliable. The proposed system can accomplish both objectives while making program management much more simple and creating savings in time, effort, storage and paper and is applicable to most funding agencies in federal and state government.

Benefits:

As noted earlier, today a typical government program manager does not have timely access to project information that he or she is currently funding. The ability to access project information does not exist or exists on a piecemeal basis with a time delay for project participants to manually integrate the data. A virtual program management process will include critical functions from the current process such as: proposal submission, budget allocation, and project tracking, but these functions will be distributed to the appropriate sites. This system will provide access to the latest research accomplishments for the program manager and if appropriate, for the public. Since the graphical user interface for the system will be developed using WWW protocols, a program manager should be able access information in the system in the office or while on travel, if a connection to the network is available.

The proposed Public Key Server will be completed and distributed to major ESnet sites by July of 1996. Widespread use of the server will take longer to accomplish because of the necessary societal changes required in individual behavior. Future work will focus on integrating the Server into mailers and other applications to facilitate its use.

20h. RELATIONSHIPS TO OTHER PROJECTS

See Technical Progress (20f.).

20j. EXPLANATION OF MILESTONES

See Future Accomplishments (20g.) for an explanation of the milestones.