Data warehouse stream view update with multiple streaming.

The main objective of data warehousing is to store information representing an integration of base data from single or multiple data sources over an extended period of time. To provide fast access to the data, regardless of the availability of the data source, data warehouses often use materialized views. Materialized views are able to provide aggregation on some attributes to help Decision Support Systems. Updating materialized views in response to modifications in the base data is called materialized view maintenance. In some applications, for example, the stock market and banking systems, the source data is updated so frequently that we can consider them as a continuous stream of data. To keep the materialized view updated with respect to changes in the base tables in a traditional way will cause query response times to increase. This thesis proposes a new view maintenance algorithm for multiple streaming which improves semi-join methods and hash filter methods. Our proposed algorithm is able to update a view which joins two base tables where both of the base tables are in the form of data streams (always changing). By using a timestamp, building updategrams in parallel and by optimizing the joining cost between two data sources it can reduce the query response time or execution time significantly.Dept. of Computer Science. Paper copy at Leddy Library: Theses & Major Papers - Basement, West Bldg. / Call Number: Thesis2005 .A336. Source: Masters Abstracts International, Volume: 44-03, page: 1391. Thesis (M.Sc.)--University of Windsor (Canada), 2005