Abstract
Profile matching is a key problem for webcasting systems. In such a system, each user has a “personal” profile. Each information document is matched with profiles in a profile-database, and pushed only to those users whose profiles match the content of the document. We present an efficient and scalable parallel profile matching algorithm which can handle a large subscription volume and diversity of information content. Our algorithm automatically partitions the profile database for load balancing and minimizes the interaction among processors in parallel profile matching. It has a dynamic load balancing mechanism for handling profile updates. We describe the implementation of our algorithm in the context of the Grand Central Station (GCS) project at the IBM Almaden Research Center. The initial performance evaluation indicates that parallel profile matching can be scaled up gracefully via dynamic adaptation.
Part of the work is done while at IBM Almaden Research Center.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
References
PointCast Inc., 1997, http://www.pointcast.com/
BackWeb Inc., 1997, http://www.backweb.com/
DataChannel Inc., 1997, http://www.datachannel.com/
Marimba Inc., 1997, http://www.marimba.com/
T.W. Yan, H. Garcia-Molina, A tool for wide-area information dissemination, Proceedings of the 1995 USENIX Technical Conference, pages 177–86, 1995.
T.W. Yan, H. Garcia-Molina, Structures for selective dissemination of information under the boolean model. ACM Transactions on Database Systems, 19(2):332–64, 1994.
Information on the Fast Track, IBM Research Magazine, Vol. 35, No.3, 1997: 18–21.
IBM: All searches start at Grand Central, Network World, Nov. 11, 1997, 1–2.
TheHarvest InformationDiscoveryandAccessSystem,1996, http://harvest.transarc.com/
C.M. Bowman, P.B. Danzig, D.R. Hardy, U. Manber and M.F. Schwartz, The Harvest information discovery and access system, Computer Networks and ISDN Systems 28 (1995) pp. 119–125.
S.-H. Teng. Greedy algorithms for low energy and mutually distant sampling. J. Algorithms, 1998.
S.-H. Teng. Coarsening, sampling, and smoothing: elements of the multilevel method. The IMA Volumes in Mathematics and Its Applications, R. Schreiber ed. Springer-Verlag, 1998.
R.L. Graham. Bounds for certain multiprocessor anomalies, Bell System Technical Journal(1966) 1563–1581.
Extensible Markup Language (XML), 1997, http://www.w3.org/XML/
K. Shoens, A. Luniewski, P. Schwarz, J. Stamos, J. Thomas, The Rufus system: information organization for semi-structured data, Proceedings of 19th International Conference on Very Large Data Bases, Dublin, Ireland, Aug. 1993.
S. Wu, U. Manber, A fast algorithm for multi-pattern searching, Technical Report TR94-17, Department of Computer Science, University of Arizona, Tucson, May 1994.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1998 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Eichstaedt, M., Lu, Q., Teng, SH. (1998). Parallel profile matching for large scale webcasting. In: Ferreira, A., Rolim, J., Simon, H., Teng, SH. (eds) Solving Irregularly Structured Problems in Parallel. IRREGULAR 1998. Lecture Notes in Computer Science, vol 1457. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0018524
Download citation
DOI: https://doi.org/10.1007/BFb0018524
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-64809-3
Online ISBN: 978-3-540-68533-3
eBook Packages: Springer Book Archive