Howard Lee Morgan

Optimal program and data locations in computer networks

An optimization procedure for the allocation of program and data files in a computer network is presented. This algorithm takes into account the dependencies between files and programs such as occur in real heterogeneous computer networks. Insights into whether or not to convert programs from one computer to another can also be gained from the model. A search procedure for the file location problem is described, along with an example and a possible application of the model.

Optimizing distributed data bases: a framework for research

The interaction between computers and communication technology has steadily developed in recent years bringing with it the creation of large computer communication complexes. Earlier computer communication systems were generally focused around a single large computer installation. Although a strong argument can still be made for serving distributed users with a centralized system, we now witness an increasing number of systems in which information processing and storage functions are distributed among several computers. The idea is to distribute the resources (programs, data, computing hardware, etc.) of each computer site to a widely spread community of users. Various factors may favor a shift toward computer networks, especially with similarly structured organizations which are independent but can be motivated to share. The most well-known example is the network developed under the sponsorship of the Advanced Research Project Agency (ARPA), that links independent research organizations, all involved in solving closely related problems. The libraries of specialized software and data at each research center, make it attractive to link the computers of these organizations. Clearly, access to specialized data bases is a major feature of the networks and the cost incurred by such an access is an important consideration in determining the economic viability of this kind of resource sharing.

A Dynamic Optimization Model for Distributed Databases

We present a dynamic programming formulation for solving the problem of locating files in heterogeneous, distributed computer networks. The model presented permits the access rates for information from files to vary over time, takes into account the fact that programs accessing files may only be permitted at certain nodes, and solves this general T-period problem. Special features are used to limit the growth of the state space so that a computationally feasible algorithm results. Finally, we present an example to illustrate the use of the algorithm.

Dynamic restructuring of databases with generation data structures

Most logical database restructuring schemes require a complete pass through the database for reformatting. Our approach is to leave the database in situ and to permit a mixture (several generations) of structures to co-exist. Each generation of structure is described in a Generation Data Structure Schema, which has a generic structure of its own. A Restructuring Data Definition Language is proposed for describing the evolution from one schema to the next. Steps toward implementation are discussed.

Display facilities for DSS support: the DAISY approach

The most complicated mathematical models, the most comprehensive data bases or the most detailed instructions are all ineffective as decision support devices unless their methods, information and results can be effectively communicated to the manager who must make the decisions. During the past 18 months, we have been engaged in the design and implementation of DAISY, a Decision Aiding Information System.5.8 Within the framework of this system, we have attempted to provide a sophisticated display system with which the manager and the decision aids can interact. This paper describes the major concepts behind this display system, and discusses the specific interactions between the other decision support facilities present in DAISY and the display system. DAISY is being developed with the support of the Office of Naval Research as part of its Operational Decision Aids Project. DAISY is meant to help tactical level managers plan and carry out missions with the assistance of a set of state-of-the-art decision aids. These include various mathematical models, access to a large, distributed data base stored on heterogeneous computers and a dynamic checklist facility to assist in remembering all of the various details in a complex decision process. In addition to the miltary context, scenarios are under development for such diverse tasks as corporate merger and acquisition planning, consolidated budgeting tasks and certain office processes. Conceptually, DAISY is mainly a person/machine interface which sits between a user on one side, and a set of decision-aiding tools on the other. These tools need not all be written in the same computer language, nor need they reside on a common computer. At the present time, decision aids (models or data base systems) written in FORTRAN, LISP, APL and POP-2 are all accessed through the DAISY system, which is itself programmed mainly in the POP-2 language on the Wharton Schools PDP-10 computer. The key concept behind the display facilities in D/~ISY is that of multiple concurrent logical terminals being displayed to the user simultaneously. If one examines a manager in the process-of decision making, one would usually find a large number of items in his or her field of view. Maps, tables, blank paper, books, procedure manuals, would very likely all be spread about on a work surface. A nearby blackboard or bulletin board would be covered with idea notation. On the other hand, if one examines the typical computer environment provided by decision support systems, one finds a single terminal-perhaps a sophisticated graphics device--displaying essentially only one type of information at a time. When we began to program the interaction facilities in DAISY, we decided that to aid the manager, the same multiplicity of inputs and outputs had to be provided. The next section discusses the window approach at a conceptual level. Then, some of the other features, including multimedia, graphics and input devices are presented. Some additional examples of how this approach can be used for specific situations are shown. Finally, a discussion of how these features really aid in DSS construction and use is given. An appendix details the features available in the window package.

Office automation project: a research perspective

This paper attempts to place some perspective on the research and developments going on in office automation. It describes the functions which can be assisted by computers, and indicates where more research may be needed. A brief description of the Office Automation Project at the Wharton School is provided. The systems being developed include word processing, electronic mail, decision aiding technology, and integration with various databases. This effort is compared with some of the other, complementary research projects in office automation under way around the country.

The people/promise tracker: preliminary description

Abstract : True automation in the office comes when the initiation of activities is aided by the system rather than the system merely serving as a passive tool to assist when the person asks. In this vein, it has been built upon the SCOOP and SEED database systems an office automation system which keeps track of the promises made to people, and vice versa.

Natural language interfaces

The panel will focus on the different approaches to creating natural language interfaces to DBMS. Of particular interest is the manner in which database definition information, user feedback dialogues, storage of the lexicon and database path selection are handled. The panelists have all been involved in implementing such interfaces.

SIGBDR (Panel Session)

Many organizations today face a unique set of problems and opportunities because they operate across international boundaries. To some extent, the growth of the multinational organization has been made possible by electronic communications. Now, more and more of these organizations are using computers to tie together their far-flung operations. The planning, organizing, and supporting of information systems for the multinational will be the focus of this session.

Multinational information systems (Panel Discussion)

Many organizations today face a unique set of problems and opportunities because they operate across international boundaries. To some extent, the growth of the multinational organization has been made possible by electronic communications. Now, more and more of these organizations are using computers to tie together their far-flung operations. The planning, organizing, and supporting of information systems for the multinational will be the focus of this session.