Query Planning in the Presence of Overlapping Sources

  • Jens Bleiholder
  • Samir Khuller
  • Felix Naumann
  • Louiqa Raschid
  • Yao Wu
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3896)

Abstract

Navigational queries on Web-accessible life science sources pose unique query optimization challenges. The objects in these sources are interconnected to objects in other sources, forming a large and complex graph, and there is an overlap of objects in the sources. Answering a query requires the traversal of multiple alternate paths through these sources. Each path can be associated with the benefit or the cardinality of the target object set (TOS) of objects reached in the result. There is also an evaluation cost of reaching the TOS.

We present dual problems in selecting the best set of paths. The first problem is to select a set of paths that satisfy a constraint on the evaluation cost while maximizing the benefit (number of distinct objects in the TOS). The dual problem is to select a set of paths that satisfies a threshold of the TOS benefit with minimal evaluation cost. The two problems can be mapped to the budgeted maximum coverage problem and the maximal set cover with a threshold. To solve these problems, we explore several solutions including greedy heuristics, a randomized search, and a traditional IP/LP formulation with bounds. We perform experiments on a real-world graph of life sciences objects from NCBI and report on the computational overhead of our solutions and their performance compared to the optimal solution.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Khuller, S., Moss, A., Naor, J.S.: The budgeted maximum coverage problem. Inf. Process. Lett. 70, 39–45 (1999)MATHCrossRefMathSciNetGoogle Scholar
  2. 2.
    Lacroix, Z., Murthy, H., Naumann, F., Raschid, L.: Links and paths through life sciences data sources. In: Rahm, E. (ed.) DILS 2004. LNCS (LNBI), vol. 2994, pp. 203–211. Springer, Heidelberg (2004)CrossRefGoogle Scholar
  3. 3.
    Mihaila, G., Naumann, F., Raschid, L., Vidal, M.E.: A data model and query language to explore enhanced links and paths in life science sources. In: Proceedings of the ACM SIGMOD Workshop on The Web and Databases, WebDB (2005)Google Scholar
  4. 4.
    Raschid, L., Vidal, M.E., Cardenas, M., Marquez, N., Wu, Y.: Challenges of navigational queries: Finding best paths in graphs. Technical report, University of Maryland (2005)Google Scholar
  5. 5.
    Khuller, S., Raschid, L., Wu, Y.: LP randomized rounding for maximum coverage problem and minimum set cover with threshold problem. Technical report, University of Maryland (2005)Google Scholar
  6. 6.
    Motwani, R., Raghavan, P.: Randomized algorithms. Cambridge University Press, Cambridge (1995)MATHGoogle Scholar
  7. 7.
    Goos, G.: Vorlesungen über Informatik - Paralleles Rechnen und nicht-analytische Lösungsverfahren, vol. 4. Springer, Berlin (1998)Google Scholar
  8. 8.
    Gruser, J.R., Raschid, L., Zadorozhny, V., Zhan, T.: Learning response time for websources using query feedback and application in query optimization. VLDB Journal 9, 18–37 (2000)CrossRefGoogle Scholar
  9. 9.
    Nie, Z., Kambhampati, S.: A frequency-based approach for mining coverage statistics in data integration. In: Proceedings of the International Conference on Data Engineering (ICDE), pp. 387–398 (2004)Google Scholar
  10. 10.
    Polyzotis, N., Garofalakis, M.: Structure and value synopses for XML data graphs. In: Proc. of the Int. Conf. on Very Large Databases, VLDB (2002)Google Scholar
  11. 11.
    Selinger, P., Astrahan, M., Chamberlin, D., Lorie, R., Price, T.: Access path selection in a relational database management system. In: Proce. of the ACM Int. Conf. on Management of Data (SIGMOD), Boston, MA, pp. 23–34 (1979)Google Scholar
  12. 12.
    Stillger, M., Lohman, G.M., Markl, V., Kandil, M.: LEO - DB2’s LEarning Optimizer. In: Proc. of the Int. Conf. on Very Large Databases (VLDB), Rome, Italy, pp. 19–28 (2001)Google Scholar
  13. 13.
    Kossmann, D.: The state of the art in distributed query processing. ACM Computing Surveys 32, 422–469 (2000)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Jens Bleiholder
    • 1
  • Samir Khuller
    • 2
  • Felix Naumann
    • 1
  • Louiqa Raschid
    • 2
  • Yao Wu
    • 2
  1. 1.Humboldt-Universität zu BerlinGermany
  2. 2.University of MarylandCollege ParkUSA

Personalised recommendations