Off-campus UMass Amherst users: To download dissertations, please use the following link to log into our proxy server with your UMass Amherst user name and password.
Non-UMass Amherst users, please click the view more button below to purchase a copy of this dissertation from Proquest.
(Some titles may also be available free of charge in our Open Access Dissertation Collection, so please check there first.)
Transaction recovery in databases and beyond
Abstract
Recovery is an essential part of databases and most computer systems, because it enables a system to exhibit good behavior in the face of failures. Although there is a considerable amount of work, both academic and industrial, on supporting recovery, we believe that much remains to be done both in terms of broadening the domain to include novel applications and platforms, and to make recovery easier to understand. This dissertation advances the state of the art in recovery by providing a simple yet powerful framework to characterize failure liveness (recoverability) properties in systems. The framework is based on ingredients common to a broad set of systems, and by introducing abstraction to the description of recovery, it enables layered decomposition of the recovery aspects of systems. This dissertation also contributes to the understanding, design, and implementation of the specific example systems (electronic commerce, mobile, database) to which we applied it. Our framework FL and its associated methodology are based on the premise that a wide range of systems, for which explicit consideration of failure and recovery is important, do share some essential ingredients. We have codified those ingredients as abstract elements: guarantees and protocols. We have used protocols and guarantees to specify and reason about the recovery properties of systems, broadly understood as those related to a system's ability to make progress in spite of failures, i.e., its liveness with respect to failures. The specification and reasoning capabilities of our framework apply to systems which operate in a loosely defined transactional nature, i.e., systems which: (i) manage valuable transactional data, representing money, goods, or other resources; (ii) require robust behavior in the face of failures, i.e., require support for recovery; (iii) may be formed by a distributed aggregation of heterogenous and autonomous components.
Subject Area
Computer science
Recommended Citation
Pedregal Martin, Cristobal, "Transaction recovery in databases and beyond" (2002). Doctoral Dissertations Available from Proquest. AAI3039385.
https://scholarworks.umass.edu/dissertations/AAI3039385