|
|
|
@inproceedings{DBLP:conf/vldb/HuaL90,
author = {Kien A. Hua and
Chiang Lee},
editor = {Dennis McLeod and
Ron Sacks-Davis and
Hans-J{\"o}rg Schek},
title = {An Adaptive Data Placement Scheme for Parallel Database Computer
Systems},
booktitle = {16th International Conference on Very Large Data Bases, August
13-16, 1990, Brisbane, Queensland, Australia, Proceedings},
publisher = {Morgan Kaufmann},
year = {1990},
isbn = {1-55860-149-X},
pages = {493-506},
ee = {db/conf/vldb/HuaL90.html},
crossref = {DBLP:conf/vldb/90},
bibsource = {DBLP, http://dblp.uni-trier.de}
}
The capacity and performance of database management system (DBMS) using a conventional (von Newmann-type) computer are limited by the total I/O channel bandwidth, the aggregate processing power, and the amount of main memory. With the advent of micro-processor, memory, and communication technology, it is econominally feasible to develop a parallel database computer system. The parallel processing techniques are employed to utilize the available resources in a coordinated fashion to solve the DBMS capacity and performance problems. Relations in such an environment are declustered into fragments and spreaded across computers. To achieve the optimal performance in data processing, it is essential for each computer to have a perfectly balanced load (i.e., identical amount of data). However, fragment sizes may vary due to insertions to and deletions from a relation. To retain good performance, the system needs to periodically rebalance data loads among the computers.
In this paper, we present an adaptive data placement scheme which balances computer work loads during query processing. The entire scheme is built on top of the popular grid file structure (but not limited to grid file). The adaptivity of the scheme and its relevant features are discussed. The cost of load rebalancing is estimated. The result shows that under our assumptions, it is always beneficial to perform load rebalancing before performing a join on skewed data.
Copyright © 1990 by the VLDB Endowment. Permission to copy without fee all or part of this material is granted provided that the copies are not made or distributed for direct commercial advantage, the VLDB copyright notice and the title of the publication and its date appear, and notice is given that copying is by the permission of the Very Large Data Base Endowment. To copy otherwise, or to republish, requires a fee and/or special permission from the Endowment.
