JOQRS: Parallel Query Optimization
The goal of the project is to develop techniques for the optimization of
SQL queries for parallel machines.
The project is so named since we view query optimization as
consisting of two phases, JOQR (Join Ordering and Query Rewrite)
followed by Scheduling . The JOQR phase transforms a query to minimize
total cost. The Scheduling phase extracts parallelism and
schedules machine resources to minimize response time.
Since parallel execution requires communicating data across processors,
our focus is on developing algorithms that incorporate the trade-off
between benefitting from parallelism and paying for communication.
Participants: Chandra Chekuri (Stanford), Susanne Englert (Tandem),
Ray Glasstone (Tandem),
Waqar Hasan (Stanford and HP),
Rajeev Motwani (Stanford),
Gio Wiederhold
(Stanford)
Support:
Financial support for the project is provided by
Hewlett-Packard Company.
Tandem Computers has provided access to a K1000 System and NonStop SQL/MP.
For more information contact Waqar Hasan (hasan@cs.stanford.edu).
Communication sure puts a dent in your speed !
Publications:
-
W. Hasan:
Optimization of SQL Queries for Parallel Machines,
PhD Thesis.
-
W. Hasan and R. Motwani:
Coloring Away Communication in Parallel
Query Optimization,
Proceedings of the 21st Intl
Conf on Very Large Data Bases,
Zurich, Switzerland, 1995, To Appear.
-
S. Englert, R. Glasstone and W. Hasan:
Parallelism and its Price: A Case Study of NonStop SQL/MP,
-
C. Chekuri, W. Hasan and R. Motwani:
Scheduling Problems in Parallel Query Optimization,
Proceedings of the ACM Symposium on Principles of Database Systems,
San Jose,
1995
-
W. Hasan and R. Motwani:
Optimization Algorithms for Exploiting the
Parallelism-Communication Tradeoff in Pipelined
Parallelism,
Proceedings of the 20th Intl
Conf on Very Large Data Bases, pages 36-47
Santiago, Chile, Sep 1994.
-
W. Hasan and P. Valduriez:
Selected Issues in Parallel Query Optimization,
Hewlett-Packard Technical Report HPL-DTD-93-3,
Used for workgroup discussion at 5th Intl Workshop on
Foundations of Models and Languages for Data and Objects
Aigen (Austria), 1993
- S. Ganguly, W. Hasan and R. Krishnamurthy:
Query Optimization for
Parallel Execution,
Proceeding of the ACM SIGMOD Intl Conf on Management of Data, pages 9-18,
San Diego, USA, June 1992