Download:
by L. Zhu, A. M. Keller, G. Wiederhold, Liping Zhu, Arthur M. Keller, Gio Wiederhold
ftp://db.stanford.edu/pub/cstr/reports/cs/tr/91/1351/CS-TR-91-1351.pdf
Add To MetaCart
Abstract:
In this report we analyze and compare hash-join based parallel multi-join algorithms for sequenced and pipelined processing. The BBN Butterfly machine serves as the host for the performance analysis. The sequenced algorithm handles the multiple join operations in a conventional sequenced manner, except that it distributes the work load of each operation among all processors. The pipelined algorithms handle the different join operations in parallel, by dividing the processors into several groups, with the data flowing through these groups. The detailed timing tests revealed the bus/memory contention that grows linearly with the number of processors. The existence of such a contention leads to an optimal region for the number of processors, given the join operands fixed. We present the analytical and experimental formulae for both algorithms, which incorporate this contention. We discuss the way of finding an optimal point, and give the heuristics for choosing the best processor’s partition in pipelined processing. The study shows that the pipelined algorithms produce the first joined result sooner than the sequenced algorithm and need less memory to store the intermediate result. The sequenced algorithm, on the other hand, takes less time to finish the whole join operations.
Citations
|
163
|
Implementation Techniques for Main Memory Database Systems
– DeWitt, Katz, et al.
- 1984
|
|
96
|
Multiprocessor hash-based join algorithms
– DeWitt, Gerber
- 1985
|
|
77
|
Hashing methods and relational algebra operations
– Bratbergsengen
- 1984
|
|
62
|
Join and Semijoin Algorithms for a Multiprocessor Database Machine
– Valduriez, Gardarin
- 1984
|
|
46
|
Parallel algorithms for the execution of relational database operations
– Bitton, Boral, et al.
- 1983
|
|
28
|
Design and Evaluation of Parallel Pipelined Join Algorithms
– Richardson, Lu, et al.
- 1987
|
|
15
|
Database Applications of the Fetch-and-Add Instruction
– Stone
- 1984
|
|
8
|
Turby ll. Benchmarking database systems, a systematic approach
– Bitton, DeWitt, et al.
- 1983
|
|
5
|
Effective Resource Utilization for Multiprocessor Join Execution
– Murphy
- 1989
|
|
4
|
et al. Application of Hash to Data Base Machine and Its Architecture
– Kitsuregawa
- 1983
|
|
2
|
Distributing a Database for Parallel Processing is NP-hard
– Du
|
|
2
|
Parallel Control Techniques for Dedicated Relational Database Engines", ICOT research center
– Itoh
|
|
2
|
Optimization of Large Join Queries: Distributions of Query Plan Costs
– Swami
- 1989
|
|
2
|
Design and Evaluation of a High-Speed Extended Relational Database Engine
– Yamane
|
|
2
|
Architecture and Algorithm for Parallel Execution of a Join Operation", CSG
– Nakay
|
|
2
|
Evaluation of Performance of the equi-join Operation on the Michigan Relational Database Machine
– Qadah
- 1984
|
|
2
|
Memory Capacity in Multiple Processor Joins
– Shultz
|
|
2
|
Notes on Databases on Parallel Computers
– Wiederhold, Keller
- 1988
|
|
2
|
The Throughput of a Precedence-Based Queuing Discipline
– Tsitsik
|
|
1
|
Benchmarking Database Systems - -4 Systematic Approach
– Bitton, DeWitt, et al.
- 1983
|
|
1
|
Parallel -4lgorithms for the Execution of Relational Database
– Bitton
- 1983
|
|
1
|
VLSI TIME/SPACE Complexity of an .4ssociative Parallel Join Module
– R
- 1986
|
|
1
|
et al., “Application of Hash to Data Base Machine and its
– Kitsuregawa
- 1983
|
|
1
|
4rchitecture and -4lgorithm for Parallel Esecution of a Join Operation”, CSG
– Nakay
- 1984
|
|
1
|
The Throughput of a Precedence-Based Queuing
– Tsitsik
|
|
1
|
VLSI TIME/SPACE Complexity of an AssociativeParallel Join Module
– R
- 1986
|
