The analysis of computer backup service: Economics, optimal capacity, and resource allocation

This dissertation develops four analytical models that provide useful results to help business managers make better decisions regarding computer systems recovery. First, the dissertation examines a computer backup service system where subscribers to a computer backup service can receive computing service replacement from a central facility, termed the hot site, owned by a service provider. It provides a guideline for business managers on whether it is cost effective to obtain the computer backup service. It also derives a formula describing the relationship of various parameters in an optimal service contract that maximizes the expected total revenues for the supplier.;Second, this dissertation finds the optimal capacity of the computer backup center to minimize the total cost to the firm, where the total cost consists of the total capacity cost of the computer backup center and the total opportunity cost of lost computing time. Third, it analyzes the allocation of computing resources within a firm where there exists information asymmetry and incentive incompatibility problems. The findings enable the headquarters to design an optimal allocation scheme such that the revelation principle will prevail. A general methodology is also presented for numerically deriving the optimal repair strategy for the computer.;Last, but not the least, this dissertation addresses internal pricing and optimal backup capacity issues for computer systems subject to failures. The findings show that the optimal long-run backup capacity should equal the firm's optimal long-run original capacity for all forms of demand functions; the firm should charge users the total marginal costs of all available computing capacities, including the cost of the backup facility; and it is imperative to obtain backup capacity for an environment where user's delay cost is high and the computer system is unreliable.