Title :
Evaluation of linear join processing trees in shared-nothing database environment
Author :
Harada, Lilian ; Akaboshi, Naoki
Author_Institution :
Fujitsu Labs. Ltd., Kawasaki, Japan
Abstract :
In this paper, aimed at understanding the effects of resource utilization in the performance of multi-way joins in shared-nothing database systems, we introduce a performance modeling for the left and right-deep linear join tree structures. The main purpose of the model is to get some insight into the behavior of the tree formats and their effects in the overall parallel processing. We present many performance results and show that the best multi-way join processing performance is achieved when the number of active operators can be adjusted in order to maximize the overlap of the disk and network transfers
Keywords :
database management systems; file organisation; query processing; tree data structures; active operators; linear join processing trees; linear join tree structures; multi-way joins; multiway join processing performance; network transfers; performance modeling; resource utilization; shared-nothing database environment; tree formats; Aggregates; Analytical models; Bandwidth; Database systems; Microprocessors; Multiprocessing systems; Parallel processing; Prototypes; Resource management; Tree data structures;
Conference_Titel :
Computing and Information, 1993. Proceedings ICCI '93., Fifth International Conference on
Conference_Location :
Sudbury, Ont.
Print_ISBN :
0-8186-4212-2
DOI :
10.1109/ICCI.1993.315338
