Query language-based inverses of schema mappings: semantics, computation, and closure properties
- 241 Downloads
- 3 Citations
Abstract
The inversion of schema mappings has been identified as one of the fundamental operators for the development of a general framework for metadata management. During the last few years, three alternative notions of inversion for schema mappings have been proposed (Fagin-inverse (Fagin, TODS 32(4), 25:1–25:53, 2007), quasi-inverse (Fagin et al., TODS 33(2), 11:1–11:52, 2008), and maximum recovery (Arenas et al., TODS 34(4), 22:1–22:48, 2009)). However, these notions lack some fundamental properties that limit their practical applicability: most of them are expressed in languages including features that are difficult to use in practice, some of these inverses are not guaranteed to exist for mappings specified with source-to-target tuple-generating dependencies (st-tgds), and it has been futile to search for a meaningful mapping language that is closed under any of these notions of inverse. In this paper, we develop a framework for the inversion of schema mappings that fulfills all of the above requirements. It is based on the notion of \({\mathcal{C}}\) -maximum recovery, for a query language \({\mathcal{C}}\), a notion designed to generate inverse mappings that recover back only the information that can be retrieved with queries in \({\mathcal{C}}\). By focusing on the language of conjunctive queries (CQ), we are able to find a mapping language that contains the class of st-tgds, is closed under CQ-maximum recovery, and for which the chase procedure can be used to exchange data efficiently. Furthermore, we show that our choices of inverse notion and mapping language are optimal, in the sense that choosing a more expressive inverse operator or mapping language causes the loss of these properties.
Keywords
Inverses Schema mappings Metadata management Closure properties Conjunctive queries Data exchange Data integrationPreview
Unable to display preview. Download preview PDF.
Supplementary material
References
- 1.Arenas, M., Barceló, P., Fagin, R., Libkin, L.: Locally consistent transformations and query answering in data exchange. In: PODS, pp. 229–240 (2004)Google Scholar
- 2.Arenas M., Pérez J., Reutter J.L., Riveros C.: Composition and inversion of schema mappings. SIGMOD Rec. 38(3), 17–28 (2009)CrossRefGoogle Scholar
- 3.Arenas M., Pérez J., Reutter J.L., Riveros C.: Inverting schema mappings: bridging the gap between theory and practice. PVLDB 2(1), 1018–1029 (2009)Google Scholar
- 4.Arenas M., Pérez J., Riveros C.: The recovery of a schema mapping: bringing exchanged data back. TODS 34(4), 22:1–22:48 (2009)CrossRefGoogle Scholar
- 5.Arenas, M., Pérez, J., Reutter, J.L.: Data exchange beyond complete data. In: PODS, pp. 83–94 (2011)Google Scholar
- 6.Arocena, P., Fuxman, A., Miller, R.J.: Composing local-as-view mappings: closure and applications. In: ICDT, pp. 209–218 (2010)Google Scholar
- 7.Barceló P.: Logical foundations of relational data exchange. SIGMOD Rec. 38(1), 49–58 (2009)CrossRefGoogle Scholar
- 8.Bernstein, P.: Applying model management to classical meta data problems. In: CIDR (2003)Google Scholar
- 9.Bernstein, P., Melnik, S.: Model management 2.0: manipulating richer mappings. In: SIGMOD, pp. 1–12 (2007)Google Scholar
- 10.Chandra, A.K., Merlin, P.M.: Optimal implementation of conjunctive queries in relational data bases. In: STOC, pp. 77–90 (1977)Google Scholar
- 11.Fagin R.: Inverting schema mappings. TODS 32(4), 25:1–25:53 (2007)CrossRefGoogle Scholar
- 12.Fagin R., Kolaitis P.G., Miller R.J., Popa L.: Data exchange: semantics and query answering. TCS 336(1), 89–124 (2005)MathSciNetMATHCrossRefGoogle Scholar
- 13.Fagin, R., Kolaitis, P.G., Nash, A., Popa, L.: Towards a theory of schema-mapping optimization. In: PODS, pp. 33–42 (2008)Google Scholar
- 14.Fagin R., Kolaitis P.G., Popa L., Tan W.C.: Composing schema mappings: second-order dependencies to the rescue. TODS 30(4), 994–1055 (2005)CrossRefGoogle Scholar
- 15.Fagin R., Kolaitis P.G., Popa L., Tan W.C.: Quasi-inverses of schema mappings. TODS 33(2), 11:1–11:52 (2008)CrossRefGoogle Scholar
- 16.Fagin R., Kolaitis P.G., Popa L., Tan W.C.: Reverse data exchange: coping with nulls. ACM Trans. Database Syst. 36(2), 11 (2011)CrossRefGoogle Scholar
- 17.Fagin R., Nash A.: The structure of inverses in schema mappings. J. ACM 57(6), 31 (2010)MathSciNetCrossRefGoogle Scholar
- 18.Fuxman, A., Kolaitis, P.G., Miller, R.J., Tan, W.C.: Peer data exchange. In: PODS, pp. 160–171 (2005)Google Scholar
- 19.de Giacomo, G., Lembo, D., Lenzerini, M., Rosati, R.: On reconciling data exchange, data integration, and peer data management. In: PODS, pp. 133–142 (2007)Google Scholar
- 20.Halevy A.Y., Ives Z., Madhavan J., Mork P., Suciu D., Tatarinov I.: The piazza peer data management system. IEEE Trans. Knowl. Data Eng. 16(7), 787–798 (2004)CrossRefGoogle Scholar
- 21.Hell P., Nes̆etr̆il J.: Graphs and Homomorphisms. Oxford University Press, USA (2004)MATHCrossRefGoogle Scholar
- 22.Kolaitis, P.G.: Schema mappings, data exchange, and metadata management. In: PODS, pp. 61–75 (2005)Google Scholar
- 23.Lenzerini, M.: Data integration: a theoretical perspective. In: PODS, pp. 233–246 (2002)Google Scholar
- 24.Madhavan, J., Halevy, A.Y.: Composing mappings among data sources. In: VLDB, pp. 572–583 (2003)Google Scholar
- 25.Maier D., Mendelzon A., Sagiv Y.: Testing implications of data dependencies. TODS 4(4), 455–469 (1979)CrossRefGoogle Scholar
- 26.Melnik, S.: Generic Model Management: concepts and Algorithms. Lecture Notes in Computer Science, vol. 2967. Springer, Berlin (2004)Google Scholar
- 27.Melnik, S., Adya, A., Bernstein, P.A.: Compiling mappings to bridge applications and databases. ACM Trans. Database Syst. 33(4), (2008)Google Scholar
- 28.Melnik, S., Bernstein, P.A., Halevy, A.Y., Rahm, E.: Supporting executable mappings in model management. In: SIGMOD, pp. 167–178 (2005)Google Scholar
- 29.ten Cate, B., Kolaitis, P.G.: Structural characterizations of schema-mapping languages. In: ICDT, pp. 63–72 (2009)Google Scholar