ColloquiaReminder: Talk by Himanshu Gupta, Stanford University, Wed. March 28

[Margery Ishmael]

Wednesday, March 28 at 2:30 p.m. in Ryerson 251

Himanshu Gupta, Department of Computer Science, Stanford University

"Selection and Maintenance of Views in a Data Warehouse"

A data warehouse is a repository (database) that integrates
information extracted from various remote data sources, with the
purpose of efficiently supporting decision support queries.  The
information derived from the data sources and stored at the warehouse
is in the form of database tables, referred to as materialized views.
In order to keep a materialized view consistent with the data at
sources, the view needs to be incrementally maintained.  The two core
issues that arise in the design of a data warehouse are: (i) selection
of views to materialize, and (ii) incremental maintenance of
materialized views.  My doctoral dissertation addressed these design
issues and presented comprehensive solutions.

For the first problem of selection of views to materialize, we
developed a theoretical framework for the general problem of selection
of views in a data warehouse. Given a set of queries to be supported,
the view selection problem is to select a set of views to materialize
minimizing the query response time under some resource constraint.
For different resource constraints and settings, we designed
approximation algorithms that provably return a set of views having a
query response time within a constant factor of the optimal.

As the data in the information sources changes, the materialized views
need to be kept up to date. Traditional maintenance algorithms maintain
views incrementally by propagating insertions and/or deletions from
the base relations to the views through intermediate subexpressions.
We developed a change-table technique, that propagates
``change-tables'' through subexpressions for incremental maintenance
of general view expressions involving aggregate and outerjoin
operators.  We showed that the developed change-table technique
outperforms the previously proposed techniques by orders of magnitude.
http://www-db.stanford.edu/~hgupta/

*The talk will be followed by refreshments in Ryerson 255*
-- 
Margery Ishmael
Department of Computer Science
The University of Chicago
1100 E. 58th Street
Chicago, IL. 60637

Tel. 773-834-8977  Fax. 773-702-8487