Design and implementation of the Wisconsin storage system
Software—Practice & Experience
The Crystal Multicomputer: Design and Implementation Experience
IEEE Transactions on Software Engineering
Query optimization in star computer networks
ACM Transactions on Database Systems (TODS)
Optimization of query evaluation algorithms
ACM Transactions on Database Systems (TODS)
Decomposition—a strategy for query processing
ACM Transactions on Database Systems (TODS)
Using Semi-Joins to Solve Relational Queries
Journal of the ACM (JACM)
Optimizing the performance of a relational algebra database interface
Communications of the ACM
Evaluation of database access paths
SIGMOD '78 Proceedings of the 1978 ACM SIGMOD international conference on management of data
Access path selection in a relational database management system
SIGMOD '79 Proceedings of the 1979 ACM SIGMOD international conference on Management of data
Hashing Methods and Relational Algebra Operations
VLDB '84 Proceedings of the 10th International Conference on Very Large Data Bases
Benchmarking Database Systems A Systematic Approach
VLDB '83 Proceedings of the 9th International Conference on Very Large Data Bases
Performance Analysis of Database Systems
Performance Evaluation: Origins and Directions
In-Network join processing for sensor networks
APWeb'06 Proceedings of the 8th Asia-Pacific Web conference on Frontiers of WWW Research and Development
| Hi-index | 0.00 |
This paper presents some experimental results on the performance of distributed join algorithms in a local network. Eight different join methods have been implemented in an experimental distributed system, the Crystal multicomputer, and tested for join queries with a variety of relation sizes, join selectivities, and join column value distributions. The results obtained indicate that pipelined join methods outperform sequential methods over a wide range of join queries. It was also found that the communications costs in a local network environment are not a dominant factor with respect to performance, and that shipping a whole relation from one site to another is not an unreasonahle decison as long as it is done in the framework of pipelined algorithms. Two pipelined variants of a centralized nested loops join algorithm (with an index) were found to perform well for many of the queries tested.