2008 Second UKSIM European Symposium on Computer Modeling and Simulation Availability and Accuracy of Distributed Web Crawlers: A Model-Based Evaluation September 08-September 10 ISBN: 978-0-7695-3325-4
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/EMS.2008.55
Distributed Web crawlers are extensively used for Web mining nowadays, but their accuracy, dependability and other operational measures have not been fully studied. Distributed Web crawlers are costly and require careful selection of configuration parameters. It is important to have some estimation about the performance, dependability and accuracy of a Web crawler. This paper presents a model-based evaluation of the accuracy and availability of a distributed Web crawler whose architecture is based on UbiCrawler. Stochastic activity networks are used for modelling the crawler. Accuracy and availability of the Web crawler are formally defined, and the effects of environmental failure rates on crawling nodes and on the availability of the whole system are discussed.
Index Terms:
Distributed Web Crawlers, Web Crawlers, Availability, Accuracy, SAN, Stochastic Activity Networks, Modeling, Mobius
Citation:
Mitra Nasri, Saeed Shariati, Mohsen Sharifi, "Availability and Accuracy of Distributed Web Crawlers: A Model-Based Evaluation," ems, pp.453-458, 2008 Second UKSIM European Symposium on Computer Modeling and Simulation, 2008 Usage of this product signifies your acceptance of the Terms of Use. | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||