View-centric Reasoning About Parallel and Distributed Computation
Saved in:
| Title: | View-centric Reasoning About Parallel and Distributed Computation |
|---|---|
| Authors: | Smith, Marc L. |
| Committee Members: | Parsons, Rebecca; Hughes, Charles E.; University of Central Florida. College of Engineering [VIAF] University of Central Florida. College of Engineering [LC] |
| Summary: | The development of distributed applications has not progressed as rapidly as its enabling technologies. In part, this is due to the difficulty of reasoning about such complex systems. In contrast to sequential systems, parallel systems give rise to parallel events, and the resulting uncertainty of the observed order of these events. Loosely coupled distributed systems complicate this even further by introducing the element of multiple imperfect observers of these parallel events. The goal of this dissertation is to advance parallel and distributed systems development by producing a parameterized model that can be instantiated to reflect the computation and coordination properties of such systems. The result is a model called paraDOS that we show to be general enough to have instantiations of two very distinct distributed computation models, Actors and tuple space. We show how paraDOS allows us to use operational semantics to reason about computation when such reasoning must account for multiple, inconsistent and imperfect views. We then extend the paraDOS model with an abstraction to support composition of communicating computational systems. This extension gives us a tool to reason formally about heterogeneous systems, and about new distributed computing paradigms such as the multiple tuple spaces support seen in Sun's JavaSpaces and IBM's T Spaces. |
| URL: | https://stars.library.ucf.edu/rtd/1986 |
| Database: | OpenDissertations |
| Abstract: | The development of distributed applications has not progressed as rapidly as its enabling technologies. In part, this is due to the difficulty of reasoning about such complex systems. In contrast to sequential systems, parallel systems give rise to parallel events, and the resulting uncertainty of the observed order of these events. Loosely coupled distributed systems complicate this even further by introducing the element of multiple imperfect observers of these parallel events. The goal of this dissertation is to advance parallel and distributed systems development by producing a parameterized model that can be instantiated to reflect the computation and coordination properties of such systems. The result is a model called paraDOS that we show to be general enough to have instantiations of two very distinct distributed computation models, Actors and tuple space. We show how paraDOS allows us to use operational semantics to reason about computation when such reasoning must account for multiple, inconsistent and imperfect views. We then extend the paraDOS model with an abstraction to support composition of communicating computational systems. This extension gives us a tool to reason formally about heterogeneous systems, and about new distributed computing paradigms such as the multiple tuple spaces support seen in Sun's JavaSpaces and IBM's T Spaces. |
|---|
