Mary Tork Roth and Peter M. Schwarz. Don't scrap it, wrap it! a wrapper architecture for legacy data sources. In Matthias Jarke, Michael J. Carey, Klaus R. Dittrich, Frederick H. Lochovsky, Pericles Loucopoulos, and Manfred A. Jeusfeld, editors, VLDB'97, Proceedings of 23rd International Conference on Very Large Data Bases, August 25-29, 1997.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....management and electronic commerce (see, for example, 13] The existence of a common set of definitions of terminology, a common ontology, makes the interoperation of different information systems much easier. This is the approach takes by many researchers in information integration, e.g. [5, 9, 11, 15]. Moreover, ontologies are useful in information retrieval, where the use of the right keywords is critical for the successful processing of a user query [7] Ontologies need to be developed and maintained just like other software parts or knowledge bases. In fact, these tasks are quite complex ....

....old concepts (faculty and visiting staff) Although faculty bears the same name in both ontologics, their meaning is different, since in Figure 4 it includes the visiting staff, too. 4 Ontology Refinement in Information Integration 4. 1 Information Integration Information integration (II) [2, 5, 9, 11, 15] seeks to bridge the gap between the user needs and the information available from different sources by introducing a meta information source, called mediator or facilitater, which provides in an integrated fashion access to information from various sources. The basic architecture of the ....

M. Tork Roth and P. Schwarz. Don't Scrap it, Wrap it! A Wrapper Architecture for Legacy Data Sources. In Proc. 23rd International Conference on Very Large Data J3ases, 1997.

....language or programming interface of the underlying source. The wrapper also defines object collections which are the targets of queries in Garlic. The wrapper further provides a description of its query processing capabilities in the form of a set of rules (encapsulated as planning methods [RS97]) Different sources may vary greatly in their query processing capabilities, and thus will provide different rules. A wrapper does not have to reflect the full query functionality of its data sources. How ever, in order for the data in that data source to be accessible through queries, some ....

....that wrappers constmct their plans using STARs. Note, however, that since Garlic does not interpret the wrapper plans (only their properties) wrappers are actually free to construct their plans however they wish, as long as the interface to Garlic is STAR like. Interested readers may consult [RS97] for the wrapper s perspective on this process. STARs provide a useful means of capturing the wrappers query capabilities, regardless of implementation. Thus, when we need to char acterize the work done in a plan by a wrapper, we will use wrapper STARs and wrapper POPs to do so. We will use ....

M. Tork Roth and P. Schwarz. Don't scrap it, wrap it! A wrapper architecture for legacy data sources. In VLDB Conf., Athens, 1997.

....links takes place, so that broken links can easily be detected or avoided, respectively. We could transform the data into another overall data model like OEM ( 5] which is the de facto standard semistructured data model ( 4] the ODMG data model ( 2] or the data model of the Garlic project ([15], 12] These models are bound to a specific implementation of an information system: Lore, an object oriented database system or the Garlic system. To facilitate online as well as offline search the specific data management system has to be installed at the student s home, which is a not ....

....of query processing. Furthermore, the data models mentioned above are too complex: Mostly, objects are not instances of (known) classes, the concepts inheritance and abstract data types as well as methods (operations) are not used. Using ODL (ODMG data model) OEM syntax or GDL (Garlic project, [15]) the specification of the data would be unnecessarily complex because only a part of it would find a use. In the area of virtual courses we have the advantage that we need only reading access to the data. Therefore we do not need a data management system. Of course, the data can be stored in a ....

Mary Tork Roth and Peter M. Schwarz. Don't scrap it, wrap it! a wrapper architecture for legacy data sources. In Matthias Jarke, Michael J. Carey, Klaus R. Dittrich, Frederick H. Lochovsky, Pericles Loucopoulos, and Manfred A. Jeusfeld, editors, VLDB'97, Proceedings of 23rd International Conference on Very Large Data Bases, August 2529,

....management and electronic commerce (see, for example, 13] The existence of a common set of definitions of terminology, a common ontology, makes the interoperation of di#erent information systems much easier. This is the approach takes by many researchers in information integration, e.g. [5, 9, 11, 15]. Moreover, ontologies are useful in information retrieval, where the use of the right keywords is critical for the successful processing of a user query [7] Ontologies need to be developed and maintained just like other software parts or knowledge bases. In fact, these tasks are quite complex ....

....and visiting sta#) Although faculty bears the same name in both ontologies, their meaning is di#erent, since in Figure 4 it includes the visiting sta#, too. 4 G. Antoniou and A. Kehagias 4 Ontology Refinement in Information Integration 4. 1 Information Integration Information integration (II) [2, 5, 9, 11, 15] seeks to bridge the gap between the user needs and the information available from di#erent sources by introducing a meta information source, called mediator or facilitator, which provides in an integrated fashion access to information from various sources. The basic architecture of the ....

M. Tork Roth and P. Schwarz. Don't Scrap it, Wrap it! A Wrapper Architecture for Legacy Data Sources. In Proc. 23rd International Conference on Very Large Data Bases, 1997.

....complex queries which involve the execution of operators from multiple function providers at di#erent sites (cycle providers) and the retrieval of data and documents from multiple data sources. In this paper we assume that all data is in a standard format (e.g. relational or XML) or wrapped [21]. Furthermore, we assume that there is a meta schema that can be used to describe all relevant properties of all services. 2.1 Query Processing Processing a query in ObjectGlobe involves four major steps (Fig. 1) lookup service optimize execute XML plug parse lookup XML query XML ....

M. Tork Roth and P. Schwarz. Don't Scrap It, Wrap It! A Wrapper Architecture for Legacy Data Sources. In Proc. of the Conf. on Very Large Data Bases (VLDB), pages 266--275, Athens, Greece, August 1997.

....object specific parameters of the corresponding hyperlink and application specific parameters that are transmitted during hyperlink processing but not contained within the hyperlink. Additional attributes of an input data object are not accessible within the HyperQuery; they are passed through. RS97] describes how more complex data sources can be queried using SQL by defining views over legacy systems. In our QueryFlow system, alternatively, a HyperQuery could consist of arbitrarily complex Java operations which have to implement the iterator interface of [Gra93] cf. Section 3.5) Figure 5 ....

M. Tork Roth and P. Schwarz. Don't Scrap It, Wrap It! A Wrapper Architecture for Legacy Data Sources. In Proc. of the Conf. on Very Large Data Bases (VLDB), pages 266--275, August 1997.

....Web wrappers [14] While it provides a powerful data extraction language (similar to recursive path expressions combined with regular expressions) the language is not tuned to XML inputs and outputs and lacks the power of XSLT templates and XPath axes and operators. The Ariadne [2] 7] Garlic [13], and TSIMMIS [4] systems are mediators that facilitate querying multiple heterogeneous sources. While Garlic and TSIMMIS support a wide range of sources, including Web sources, database systems, and file systems, Ariadne focuses on Web sources exclusively. In each system, a modeling process ....

Mary Tork Roth and Peter Schwartz. Don't Scrap It, Wrap It! A Wrapper Architecture for Legacy Data Sources. Proc. International Conference on Very Large Data Bases (VLDB), Athens, Greece, August 1997.

....results produced by each component. Schema Engine A typical session with Clio starts with the user loading one or more schemas into the system. These schemas are read from either an underlying Object Relational database, a legacy source that has been wrapped with a Garlic ObjectRelational wrapper [TS97] or from an XML file with an associated XML schema. The schemas may be legacy schemas or they may include an integrated schema produced manually or by an integration tool. The schema engine is used to augment the schema with additional constraint information, if necessary. Currently, Clio makes ....

M. Tork Roth and P. Schwarz. Don't Scrap It, Wrap It! A Wrapper Architecture for Legacy Data Sources. In Proc. of the Int'l Conf. on Very Large Data Bases (VLDB), pages 266-- 275, Athens, Greece, August 1997.

....in their operating systems. More recently, newer distributed database systems have become very popular, with most commercial systems o#ering some form of distributed query processing [24] There are also several research prototypes that provide distributed query execution, including Garlic [15, 30], DISCO [40] and Mariposa [34, 36] 2.4 Using Client Resources One of the decisions that must be made when designing a client server or distributed architecture is how to partition the query processing e#ort between the computers involved in the system. Traditional RDBMSs execute all of the ....

....in parallel with server disk resources. The hybrid approach is used in some database products such as UniSQL, application systems such as SAP R 3, and research prototypes such as ORION2 [23] This approach is also used by distributed wrapper based heterogenous systems such as DISCO [40] Garlic [15, 30], and Mariposa [34, 36] The ADMS project [31] also uses hybrid shipping, although it requires that all aggregation, duplicate elimination, and sorting be done at the client during output generation, while joins and selections can be executed either at the server or at the client accessing cached ....

Mary Tork Roth and Peter Schwarz. Don't scrap it, wrap it! A wrapper architecture for legacy data sources. In Proc. Int'l Conf. on VLDB, Athens, Greece, 1997. 24

.... local as view approach, as theoretical completeness independent of pure design decisions cannot be ascertained in a GAV approach (for an example see [30] We want to follow some of the major architectural principles shared by both the global asview approach and the local as view approach (e.g. [10, 25]) namely an architecture based on wrappers around sources that cover structural integration (i.e. the main purpose of wrappers is to translate between the query language of the integration system and the source) On top of the wrappers, we need a reasoning system that allows to rewrite queries ....

Mary Tork Roth and Peter Schwarz. Don't Scrap It, Wrap It! A Wrapper Architecture for Legacy Data Sources. Proc. VLDB 1997. 11

....using our system. The development of techniques for schema integration in a distributed and heterogeneous environment is not the target of our work because this has been addressed in other work (e.g. SL90] we assume that all data is in a standard format (e.g. relational or XML) or wrapped [RS97] Although, selling services is one of the main motiviations for the ObjectGlobe project, the system does not require a particular business model; many di erent business models can be implemented on top of ObjectGlobe. The remainder of this paper is structured as follows: Section 2 gives an ....

M. Tork Roth and P. Schwarz. Don't scrap it, wrap it! A wrapper architecture for legacy data sources. In Proc. of the Conf. on Very Large Data Bases (VLDB), pages 266275, Athens, Greece, August 1997.

....using our system. The development of techniques for schema integration in a distributed and heterogeneous environment is not the target of our work because this has been addressed in other work (e.g. 37] we assume that all data is in a standard format (e.g. relational or XML) or wrapped [33]. Although, selling services is one of the main motiviations for our project, the system does not require a particular business model; many different business models can be implemented on top of ObjectGlobe. The remainder of this paper is structured as follows: Section 2 gives an overview of ....

M. Tork Roth and P. Schwarz. Don't scrap it, wrap it! A wrapper architecture for legacy data sources. In Proc. of the Conf. on Very Large Data Bases (VLDB), pages 266--275, Athens, Greece, August 1997.

....using our system. The development of techniques for schema integration in a distributed and heterogeneous environment is not the target of our work because this has been addressed in other work (e.g. SL90] we assume that all data is in a standard format (e.g. relational or XML) or wrapped [RS97] Although, selling services is one of the main motiviations for our project, the system does not require a particular business model; many different business models can be implemented on top of ObjectGlobe. The remainder of this paper is structured as follows: Section 2 gives an overview of ....

M. Tork Roth and P. Schwarz. Don't scrap it, wrap it! A wrapper architecture for legacy data sources. In VLDB [VLD97], pages 266--275.

.... local as view approach, as theoretical completeness independent of pure design decisions cannot be ascertained in a GAV approach (for an example see [30] We want to follow some of the major architectural principles shared by both the global asview approach and the local as view approach (e.g. [10, 25]) namely an architecture based on 7 wrappers around sources that cover structural integration (i.e. the main purpose of wrappers is to translate between the query language of the integration system and the source) On top of the wrappers, we need a reasoning system that allows to rewrite ....

Mary Tork Roth and Peter Schwarz. Don't Scrap It, Wrap It! A Wrapper Architecture for Legacy Data Sources. Proc. VLDB 1997.

....using our system. The development of techniques for schema integration in a distributed and heterogeneous environment is not the target of our work because this has been addressed in other work (e.g. SL90] we assume that all data is in a standard format (e.g. relational or XML) or wrapped [RS97] Although, selling services is one of the main motiviations for the ObjectGlobe project, the system does not require a particular business model; many di#erent business models can be implemented on top of ObjectGlobe. The remainder of this paper is structured as follows: Section 2 gives an ....

M. Tork Roth and P. Schwarz. Don't scrap it, wrap it! A wrapper architecture for legacy data sources. In Proc. of the Conf. on Very Large Data Bases (VLDB), pages 266--275, Athens, Greece, August 1997.

....view and access interface to the data sets stored by each data source. The translation of the data items to the global schema is performed by either a wrapper or database gateway. Wrappers are used when integration is achieved through a mediator system, such as TSIMMIS [1] DISCO [13] or Garlic [11]. On the other hand, gateways are used when integration is realized by importing the data into a commercial DBMS, such as Oracle [6] or Informix [5] A problem with the use of middleware systems is the deployment of the applicationspecific data types and operators necessary to implement the ....

Mary Tork Roth and Peter Schwarz. Don't Scrap It, Wrap It! A Wrapper Architecture for Legacy Data Sources. In 23rd VLDB Conference, Athens, Greece, 1997.

No context found.

Mary Tork Roth and Peter M. Schwarz. Don't scrap it, wrap it! A wrapper architecture for legacy data sources. In Proceedings of the 23rd VLDB Conference, pages 266-275, Athens, Greece, August 1997. VLDB Endowment, Saratoga, Calif.

....are [Pap96] Lev96] which consider in particular the web sources. In many projects as H20, TSIMMIS, DWQ, strong attention is paid to data integration [Hull96] Hull97] Pap96] Cal99] In order to translate heterogeneous data models to a common model, some authors propose the use of wrappers [Lab97][Tork97], which encapsulate data sources and mediate between them and the rest of the system. Data transformation layer involves a wide range of transformations that have to be applied to source data, for example data quality control and data cleaning, data integration, and conversions that are necessary ....

M. Tork Roth, P. Schwarz. Don't Scrap It, Wrap It! A Wrapper Architecture for Legacy Data Sources. VLDB 1997.

.... database systems, which allow to integrate external data and functionality by the means of extensions (datablades, extenders or cartridges) or heterogeneous database systems such as # This work is supported by the German National Research Foundation DFG under contract Ke 401 7 1 1 Garlic [MS97] or Tsimmis [GMPQ 97] our approach makes it possible to place external query operators anywhere in a query evaluation plan as opposed to restricting the placement of external operations to the access level of plans. It would, for example, be possible to make our system execute a completely ....

M. Tork Roth and P. M. Schwarz. Don't scrap it, wrap it! A wrapper architecture for legacy data sources. In VLDB'97, Proc. of the Conf. on Very Large Data Bases (VLDB), pages 266--275, Athens, Greece, August 1997.

....integrate a large variety of data sources; e.g. image systems, video systems, text retrieval systems, WWW search engines, and legacy systems with specialized sets of functions. Garlic is an example of a multimedia middleware system based on an objectoriented data model. Wrappers in Garlic [RS97] describe the data in their source as objects, and Garlic refers to these objects using an OID the wrapper manufactures. This OID allows Garlic to apply methods to objects; from the OID, Garlic can determine the appropriate wrapper, and the wrapper can locate the necessary data and apply the ....

.... (These methods are typically implemented as commands in the native language or programming interface of the underlying source) The wrapper also defines object collections (the targets of queries in Garlic) Finally, the wrapper provides a description of the source s query processing capabilities [RS97] Different sources may vary greatly in their query processing capabilities. Most sources are able to scan a collection, retrieving OIDs and possibly other attributes as needed. Most multimedia sources can also order 2 Again, see [Fag96] for a more complete and general analysis of the ....

[Article contains additional citation context not shown here]

M. Tork Roth and P. Schwarz. Don't scrap it, wrap it! A wrapper architecture for legacy data sources. In Proc. of the Conf. on Very Large Data Bases (VLDB), Athens, Greece, 1997.

No context found.

Mary Tork Roth and Peter M. Schwarz. Don't scrap it, wrap it! a wrapper architecture for legacy data sources. In Matthias Jarke, Michael J. Carey, Klaus R. Dittrich, Frederick H. Lochovsky, Pericles Loucopoulos, and Manfred A. Jeusfeld, editors, VLDB'97, Proceedings of 23rd International Conference on Very Large Data Bases, August 25-29, 1997.

No context found.

Mary Tork Roth and Peter M. Schwarz. Don't scrap it, wrap it! a wrapper architecture for legacy data sources. In Matthias Jarke, Michael J. Carey, Klaus R. Dittrich, Frederick H. Lochovsky, Pericles Loucopoulos, and Manfred A. Jeusfeld, editors, VLDB'97, Proceedings of 23rd International Conference on Very Large Data Bases, August 25-29, 1997.

No context found.

Mary Tork Roth and Peter M. Schwarz. Don't scrap it, wrap it! a wrapper architecture for legacy data sources. In Matthias Jarke, Michael J. Carey, Klaus R. Dittrich, Frederick H. Lochovsky, Pericles Loucopoulos, and Manfred A. Jeusfeld, editors, VLDB'97, Proceedings of 23rd International Conference on Very Large Data Bases, August 25-29, 1997.

No context found.

Mary Tork Roth and Peter M. Schwarz. Don't scrap it, wrap it! a wrapper architecture for legacy data sources. In Matthias Jarke, Michael J. Carey, Klaus R. Dittrich, Frederick H. Lochovsky, Pericles Loucopoulos, and Manfred A. Jeusfeld, editors, VLDB'97, Proceedings of 23rd International Conference on Very Large Data Bases, August 25-29, 1997.

No context found.

M. Tork Roth and P. Schwarz. Don't scrap it, wrap it! a wrapper architecture for legacy data sources. In Proc. of the 23rd VLDB Conference, Athens, Greece, pages 266-275, 1997.

First 50 documents

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC