ParSets for parallelizing OODBMS traversals: implementation and performance

Authors:
David J. De Witt;Jeffrey F. Naughton;John C. Shafer;Shivakumar Venkataraman
Affiliations:
-;-;-;-
Venue:
PDIS '94 Proceedings of the third international conference on on Parallel and distributed information systems
Year:
1994

Citing 0
Cited 3

Run-time adaptation in river

ACM Transactions on Computer Systems (TOCS)
Critical Database Technologies for High Energy Physics

VLDB '97 Proceedings of the 23rd International Conference on Very Large Data Bases
Parallel Object Server for Fine Grained Objects

SOFSEM '00 Proceedings of the 27th Conference on Current Trends in Theory and Practice of Informatics

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper we describe the design and implementation of Par Sets, a means of exploiting parallelism in the SHORE OODBMS. We used ParSets to create and parallelize the graph traversals of OO7 OODBMS benchmark, and present speedup and scaleup results from parallel SHORE running these traversals on a cluster of commodity workstations connected by a standard ethernet. For some OO7 traversals, SHORE achieved excellent speedup and scaleup; for other OO7 traversals, only marginal speedup and scaleup occured. The characteristics of these traversals shed light on when the ParSet approach to parallelism can and cannot be applied to speed up an application.