The subject of this thesis is synchronous master-slave database replication, failover and distributed recovery for a cluster of database servers. Our replication solution allows updates on the master, and queries on ail sites. Updates of transactions are sent to ail sites before they are committed on the master. To do this, we take advantage of the semantics of group communication Systems. With failover, there is always a master in the group. Distributed recovery ensures that a site that has been introduced in the System brings its own copy of the database to be consistent with the rest of the other databases. It does this by getting all the transactions it missed from another site while it was not in the group. It is only after having successfully carried out failover and recovery mechanisms that a new site is ready to share in the load of the system.
| Identifer | oai:union.ndltd.org:LACETR/oai:collectionscanada.gc.ca:QMM.19689 |
| Date | January 2003 |
| Creators | Chouk, Mabrouk |
| Publisher | McGill University |
| Source Sets | Library and Archives Canada ETDs Repository / Centre d'archives des thèses électroniques de Bibliothèque et Archives Canada |
| Language | English |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Format | application/pdf |
| Coverage | Master of Science (School of Computer Science) |
| Rights | All items in eScholarship@McGill are protected by copyright with all rights reserved unless otherwise indicated. |
| Relation | alephsysno: 002022776, Theses scanned by McGill Library. |
Page generated in 0.0023 seconds