Parallel algorithms for computing temporal aggregates

Jose Alvin G Gendrano, Bruce C. Huang, Jim M. Rodrigue, Bongki Moon, Richard Thomas Snodgrass

Research output: Chapter in Book/Report/Conference proceedingChapter

19 Citations (Scopus)

Abstract

The ability to model the temporal dimension is essential to many applications. Furthermore, the rate of increase in database size and response time requirements has outpaced advancements in processor and mass storage technology, leading to the need for parallel temporal database management systems. In this paper, we introduce a variety of parallel temporal aggregation algorithms for a shared-nothing architecture based on the sequential Aggregation Tree algorithm. Via an empirical study, we found that the number of processing nodes, the partitioning of the data, the placement of results, and the degree of data reduction effected by the aggregation impacted the performance of the algorithms. For distributed results placement, we discovered that Time Division Merge was the obvious choice. For centralized results and high data reduction, Pairwise Merge was preferred regardless of the number of processing nodes, but for low data reduction, it only performed well up to 32 nodes. This led us to a centralized variant of Time Division Merge which was best for larger configurations having low data reduction.

Original languageEnglish (US)
Title of host publicationProceedings - International Conference on Data Engineering
PublisherInstitute of Electrical and Electronics Engineers Computer Society
Pages418-427
Number of pages10
StatePublished - 1999
EventProceedings of the 1999 15th International Conference on Data Engineering, ICDE-99 - Sydney, NSW, AUS
Duration: Mar 23 1999Mar 26 1999

Other

OtherProceedings of the 1999 15th International Conference on Data Engineering, ICDE-99
CitySydney, NSW, AUS
Period3/23/993/26/99

Fingerprint

Parallel algorithms
Data reduction
Agglomeration
Processing

ASJC Scopus subject areas

  • Software
  • Engineering(all)
  • Engineering (miscellaneous)

Cite this

Gendrano, J. A. G., Huang, B. C., Rodrigue, J. M., Moon, B., & Snodgrass, R. T. (1999). Parallel algorithms for computing temporal aggregates. In Proceedings - International Conference on Data Engineering (pp. 418-427). Institute of Electrical and Electronics Engineers Computer Society.

Parallel algorithms for computing temporal aggregates. / Gendrano, Jose Alvin G; Huang, Bruce C.; Rodrigue, Jim M.; Moon, Bongki; Snodgrass, Richard Thomas.

Proceedings - International Conference on Data Engineering. Institute of Electrical and Electronics Engineers Computer Society, 1999. p. 418-427.

Research output: Chapter in Book/Report/Conference proceedingChapter

Gendrano, JAG, Huang, BC, Rodrigue, JM, Moon, B & Snodgrass, RT 1999, Parallel algorithms for computing temporal aggregates. in Proceedings - International Conference on Data Engineering. Institute of Electrical and Electronics Engineers Computer Society, pp. 418-427, Proceedings of the 1999 15th International Conference on Data Engineering, ICDE-99, Sydney, NSW, AUS, 3/23/99.
Gendrano JAG, Huang BC, Rodrigue JM, Moon B, Snodgrass RT. Parallel algorithms for computing temporal aggregates. In Proceedings - International Conference on Data Engineering. Institute of Electrical and Electronics Engineers Computer Society. 1999. p. 418-427
Gendrano, Jose Alvin G ; Huang, Bruce C. ; Rodrigue, Jim M. ; Moon, Bongki ; Snodgrass, Richard Thomas. / Parallel algorithms for computing temporal aggregates. Proceedings - International Conference on Data Engineering. Institute of Electrical and Electronics Engineers Computer Society, 1999. pp. 418-427
@inbook{03fc48059b604a0ba9a9fcbd7c181839,
title = "Parallel algorithms for computing temporal aggregates",
abstract = "The ability to model the temporal dimension is essential to many applications. Furthermore, the rate of increase in database size and response time requirements has outpaced advancements in processor and mass storage technology, leading to the need for parallel temporal database management systems. In this paper, we introduce a variety of parallel temporal aggregation algorithms for a shared-nothing architecture based on the sequential Aggregation Tree algorithm. Via an empirical study, we found that the number of processing nodes, the partitioning of the data, the placement of results, and the degree of data reduction effected by the aggregation impacted the performance of the algorithms. For distributed results placement, we discovered that Time Division Merge was the obvious choice. For centralized results and high data reduction, Pairwise Merge was preferred regardless of the number of processing nodes, but for low data reduction, it only performed well up to 32 nodes. This led us to a centralized variant of Time Division Merge which was best for larger configurations having low data reduction.",
author = "Gendrano, {Jose Alvin G} and Huang, {Bruce C.} and Rodrigue, {Jim M.} and Bongki Moon and Snodgrass, {Richard Thomas}",
year = "1999",
language = "English (US)",
pages = "418--427",
booktitle = "Proceedings - International Conference on Data Engineering",
publisher = "Institute of Electrical and Electronics Engineers Computer Society",

}

TY - CHAP

T1 - Parallel algorithms for computing temporal aggregates

AU - Gendrano, Jose Alvin G

AU - Huang, Bruce C.

AU - Rodrigue, Jim M.

AU - Moon, Bongki

AU - Snodgrass, Richard Thomas

PY - 1999

Y1 - 1999

N2 - The ability to model the temporal dimension is essential to many applications. Furthermore, the rate of increase in database size and response time requirements has outpaced advancements in processor and mass storage technology, leading to the need for parallel temporal database management systems. In this paper, we introduce a variety of parallel temporal aggregation algorithms for a shared-nothing architecture based on the sequential Aggregation Tree algorithm. Via an empirical study, we found that the number of processing nodes, the partitioning of the data, the placement of results, and the degree of data reduction effected by the aggregation impacted the performance of the algorithms. For distributed results placement, we discovered that Time Division Merge was the obvious choice. For centralized results and high data reduction, Pairwise Merge was preferred regardless of the number of processing nodes, but for low data reduction, it only performed well up to 32 nodes. This led us to a centralized variant of Time Division Merge which was best for larger configurations having low data reduction.

AB - The ability to model the temporal dimension is essential to many applications. Furthermore, the rate of increase in database size and response time requirements has outpaced advancements in processor and mass storage technology, leading to the need for parallel temporal database management systems. In this paper, we introduce a variety of parallel temporal aggregation algorithms for a shared-nothing architecture based on the sequential Aggregation Tree algorithm. Via an empirical study, we found that the number of processing nodes, the partitioning of the data, the placement of results, and the degree of data reduction effected by the aggregation impacted the performance of the algorithms. For distributed results placement, we discovered that Time Division Merge was the obvious choice. For centralized results and high data reduction, Pairwise Merge was preferred regardless of the number of processing nodes, but for low data reduction, it only performed well up to 32 nodes. This led us to a centralized variant of Time Division Merge which was best for larger configurations having low data reduction.

UR - http://www.scopus.com/inward/record.url?scp=0032632994&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0032632994&partnerID=8YFLogxK

M3 - Chapter

SP - 418

EP - 427

BT - Proceedings - International Conference on Data Engineering

PB - Institute of Electrical and Electronics Engineers Computer Society

ER -