Efficiently updating materialized views dblp who is hannah davis dating
A data warehouse is a large data repository for the purpose of analysis and decision making in organizations.
To improve the query performance and to get fast access to the data, data is stored as materialized views (MV) in the data warehouse.
We propose an algorithm that uses key and referential constraints which reduces the total number of tuples in auxiliary views and uses idea of information sharing between these auxiliary views to further reduce number of auxiliary views.
Many issues should be studied, including how to detect changes and how to develop algorithms for warehouse maintenance that are able to balance update cost against query response time.
DB2, Oracle, SQL Server, have been successfully developed completely enough with materialized views so far.
The barrier lies in building a module that can incrementally update the materialized views automatically, which corresponds to data changes in the base tables.
From a computer science perspective, a data warehouse is a collection of materialized views derived from base relations that may not reside at the warehouse.
Therefore, a data warehouse is considered as a definer and storage of views.