Computing system reliability modeling, analysis, and optimization

Abstract

Along with the explosive development of information technology in recent decades, the concept of computing system has been widely accepted to many practical areas. To measure computing system performance, reliability has been proved as a necessary and effective metric and well studied from many different aspects by countless researches in literature. However, exponential growth in size, complexity, and functionality of the computing systems makes quantitatively modeling of system reliability difficult but even more necessary. Henceo< this dissertation investigates some important issues related to reliability modeling and analysis of various computing systems, such as reliability modeling of weighted voting systems, reliability modeling of peer-to-peer computing systems, software systems reliability and etc. With formulating these models, further research on uncertainty analysis and parameter estimation are conducted at both component level and system level. Problems of optimization and resource allocation strategies are addressed as well for better improving computing system reliability.