Updating duplicate values in distributed multidatabase systems
Shin’ichi Konomi, T. Furukawa · 2002
Redundant data for fast retrieval increase costs of updates in distributed multidatabase systems. In this paper, costs of updates are reduced by introducing further redundancies. Duplicate values-redundant data that are obtained from relationships of original values-must be updated to preserve their currency when original values are updated. To update duplicate values, relationships of original data are retrieved in order to determine which duplicate values must be updated. If a file with duplicate values satisfies the super-key condition introduced in this paper, one can perform the retrieval required for updates faster. Redundant data is added to a database site to satisfy the condition.>