Browse Theses

Computing systems that are shipped today are tested primarily at the functional level to determine if the system meets its specifications in the absence of errors. No effort is made to determine what will happen if the software or hardware does not ...

For many years, there has been tremendous interest in methods to make computation more reliable. In this thesis, we explore various techniques that can be implemented in software to help insure the correctness of the output of a program. The basic tool ...

Chime is a "distributed" parallel processing system that provides true multiprocessor semantics, including a parallel programming interface targeted for multiprocessors, in an unreliable, asynchronous distributed environment. The language support ...

This research addresses issues in simulation-based system level dependability analysis of fault-tolerant computer systems. The issues and difficulties of providing a general simulation-based approach for system level analysis are discussed and a ...

With the prevalence of database application and the widespread availability of communication networks, distributed databases systems remains an important topic in the 1990's. While distributed database systems have been the subject of extensive research,...

This dissertation introduces a formal scheme, called NEST, for fault tolerance in multiprocessor systems, its theoretical and experimental validation, and a novel technique for fault tolerance based on natural algorithmic redundancy.

NEST includes a ...

Many new software development projects employ distributed system architectures in order to increase system performance, utilization, dependability, availability and reliability. However, the potential benefits of distributed systems have not been fully ...

The design, modeling, verification, and validation of highly reliable systems has become increasingly important in recent years. These systems, particularly those which are distributed or involve real-time constraints, are extremely complex and the ...

The introduction of replication into distributed systems is necessary if they are to provide availability and fault tolerance. Although considerable research effort has been directed towards the design of replication-control protocols, replication is ...

Deadlock detection is a fundamental problem in a distributed system and has been extensively studied in the past few years. Many distributed deadlock detection/resolution algorithms have been proposed, but most of them either have not given a ...

Distributed Shared Memory (DSM) Multicomputers offer the advantages of the shared memory programming model on high-performance, scalable hardware. DSM implementations must not limit scalability or introduce performance bottlenecks. The reliability ...

This dissertation presents a new network structure called the Interleaved Cyclic Ring (ICR) for a large scale ring-based Scalable Coherent Interface (SCI) shared-memory multiprocessor system. The SCI utilizes the IEEE standard that allows for a high ...

This work introduces an automated approach to map the user's parallel program written for fault-free underlying structure onto the faulty structure. The approach tries to keep the running time invariant in the mapping. If that is not possible, the ...

The most widely used approach to building replicated, fault-tolerant services is the primary-backup approach. In this approach, the state of the service is replicated across multiple servers, with one server designated as the primary, and the rest as ...

This dissertation deals with proactive fault management, a technique for software fault tolerance, in operational software systems. The effectiveness of this technique lies in the fact that real-life software systems experience aging while in operation. ...

Asynchronous Transfer Mode (ATM) has generated new challenging problems for computer network designers. One of the challenges lies in the efficient management of ATM traffic so that the communication bandwidth of the network can be optimally utilized. ...

Design of computer systems with very high reliability and quality characteristics have become increasingly difficult and extremely challenging. A new method of determining a high testability capability of computer systems using combinatorics combined ...

In a client-server architecture, replicated copies of an object are assigned to different servers (possibly client sites) in order to increase the availability of the database in case some but not all servers fail. The aim of this work is to introduce a ...