Date of Award

1994

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Computer Science

First Advisor

Doris L. Carver

Abstract

High-speed computer networking, interactive service, and incremental growth for computing are some of the motivations for developing a distributed system. Despite the inherent benefits of a distributed system, the development of software support is more difficult for distributed systems than for sequential systems. In either case, difficulties may arise from the communication problems between two groups of people with different backgrounds trying to formulate requirements for the system. This process depends on feedback and may take many iterations to converge. Customers can usually recognize the features they need when they start using a system, which makes prototyping an important tool in requirement analysis. Many prototyping goals, objectives, and approaches are possible. Executable formal specifications are the most attractive ones. This unification of specification and prototyping by having code generators has advantages of providing consistency and prototyping at higher levels of abstraction. Thus, a methodology for executing the DOSL (Distributed Object-based Specification Language) is defined and a prototype system is developed. DOSL is extended as a new formal distributed object-oriented specification language, DOSL-II. DOSL-II is object-oriented rather than object-based, and includes class, inheritance, simple I/O, stream I/O, concurrent I/O, and new constructs for object communication.

Pages

103

Share

COinS