Quick Search:

uzh logo
Browse by:
bullet
bullet
bullet
bullet

Zurich Open Repository and Archive 

Permanent URL to this publication: http://dx.doi.org/10.5167/uzh-56413

Timko, Igor; Böhlen, Michael H; Gamper, Johann (2011). Sequenced spatiotemporal aggregation for coarse query granularities. VLDB Journal, 20(5):721-741.

[img]Published Version
PDF - Registered users only
1134Kb
[img]
Preview
Accepted Version
PDF
2806Kb

Abstract

Sequenced spatiotemporal aggregation (SSTA) is an important query for many applications of spatiotemporal databases, such as traffic analysis. Conceptually, an SSTA query returns one aggregate value for each individual spatiotemporal granule. While the data is typically recorded at a fine granularity, at query time a coarser granularity is common. This calls for efficient evaluation strategies that are granularity aware. In this paper, we formally define an SSTA operator that includes a data-to-query granularity conversion. Based on a discrete time model and a discrete 1.5 dimensional space model, we generalize the concept of time constant intervals to constant rectangles, which represent maximal rectangles in the spatiotemporal domain over which an aggregation result is constant. We propose an efficient evaluation algorithm for SSTA queries that takes advantage of a coarse query granularity. The algorithm is based on the plane sweep paradigm, and we propose a granularity aware event point schedule, termed gaEPS, and a granularity aware sweep line status, termed gaSLS. These data structures store space and time points from the input relation in a compressed form using a minimal set of counters. In extensive experiments, we show that for coarse query granularities gaEPS significantly outperforms a basic EPS that is based on an extension of previous work, both in terms of memory usage and runtime.

Item Type:Journal Article, refereed, original work
Communities & Collections:03 Faculty of Economics > Department of Informatics
DDC:000 Computer science, knowledge & systems
Language:English
Date:2011
Deposited On:10 Feb 2012 12:26
Last Modified:19 Apr 2014 15:56
Publisher:Springer
ISSN:1066-8888 (P) 0949-877X (E)
Additional Information:From the issue entitled "Special issue: Data management for mobile services". The original publication is available at www.springerlink.com
Publisher DOI:10.1007/s00778-011-0247-5
Other Identification Number:merlin-id:6168
Citations:Web of Science®
Google Scholar™

Users (please log in): suggest update or correction for this item

Repository Staff Only: item control page