Quick Starts Guide

The Simple Client quick start teaches you how to connect to JBoss Data Virtualization using either the JDBC driver or a data source. You can use the simple client quick start to test your SQL queries against any JBoss Data Virtualization instance with an active virtual database. Other quick start examples in this guide use this example as a means to test suggested queries. Refer to the individual README.md file for those quick starts in order to setup the environment correctly before executing a test query.

To run the quick start, go to the EAP_HOME/quickstarts/simpleclient directory and open the README.md file in a text editor and read the instructions it contains.

The Data Federation quick start teaches you how Red Hat JBoss Data Virtualization can federate data from different sources into one virtual database.

The quick start uses three sample data sources to illustrate this concept, namely

The virtual database to be deployed is called Portfolio. It defines a model for each data source; these will be named Accounts, MarketData and PersonalValuations, respectively. The VDB will be deployed to the Teiid Runtime, thereby making it accessible to the user application via JDBC.

To run the quick start, go to the EAP_HOME/quickstarts/dynamicvdb-datafederation directory and open the README.md file in a text editor and read the instructions it contains.

When you run the quick start’s test query, it joins data from the Accounts and MarketData models. Red Hat JBoss Data Virtualization then accesses both the relational and non-relational sources, calculates the portfolio values, and returns the results.

The dynamicvdb-datafederation/src/vdb/portfolio-vdb.xml file defines the resources that can be accessed by the client application.

In this model, the "vdb" element defines the virtual database that has a name of "Portfolio" with version "1" A "model" element represents a schema that is being integrated. This sample defines three sources. The "source" element inside the "model" element defines the "name" of the source (can be any name), the name of the translator (defines the type of the source like oracle, db2, mysql, h2, file or WS) and the "connection-jndi-name" defines the source’s JNDI name in the Red Hat JBoss EAP container.

Also note that inside the "model" elements, some "property" elements can be used to define how metadata is to be imported from the source. When you configure the quick start, the portfolio-vdb.xml is deployed to the JBoss EAP server. The portfolio-vdb.xml.dodeploy file (also referred to as a marker file, which is empty) is also deployed so that the Teiid Runtime is triggered to load the virtual database. The VDB’s data sources will then be validated for availability, and if successful, the VDB status will be set to "Active".

If the data sources are created and the VDB is deployed correctly, then you will see an information message in the JBoss EAP console that states that the VDB has been deployed and is in an "active" state. If there are any errors then you will see that the VDB was not deployed or is currently in an "inactive" state. Find the issues via the console and fix them before you proceed any further.

There are two virtual models defined in the dynamic vdb; Stocks and OtherHoldings. The Stocks virtual model contains 2 basic views. One, Stocks, to access the stock prices in the text data source and the second, StockPrices, that will join the Stocks with Products to federate the data from both data sources. The OtherHoldings contains a view that joins Acccounts with OtherHoldings to show who the holdings belongs to.

This example uses three data sources, a relational database, a CSV text file and an Excel file. You can substitute any other relational database for H2, as long as you have a suitable JDBC driver. The schema file provided at dynamicvdb-datafederation/src/teiidfiles/customer-schema.sql, is specific to H2 but you can convert it for use with other databases

In order to use a text file as the source, we need a data file which defines the data in a CSV format (comma separated values). Each data file contains column information for the table. The column information is typically defined on line 1, as header line in the file, and all the following lines contain the actual rows of data. Each line corresponds to single row. A portion of the sample file is shown below. The complete sample file is dynamicvdb-datafederation/src/teiidfiles/data/marketdata-price.txt.

SYMBOL,PRICE
IBM,83.46
RHT,11.84
BA, 44.58
ORCL,17.37

The Excel file contains sheet1, that defines a header row and subsequent data rows, similar to a text file. This example will demonstrate that excel files can be used on different operating systems, not just windows. You can use the provided data files or create your own data files. When performing the set up step, data sources will be configured.

The dynamicvdb-dataroles quick start demonstrates how you can use data roles to control access to data. (This includes read-only and read-write access.)

In this quick start, the virtual database (portfolio-vdb.xml) has these two data access roles:

To run the quick start, go to the EAP_HOME/quickstarts/dynamicvdb-dataroles directory and open the README.md file in a text editor and read the instructions it contains. Look at the application-roles.properties and application-users.properties, because you have to use the ./add-user.sh (bat) scripts (as indicated in the README).

The quick start shows use of the read-write data-role in the portfolio-vdb.xml file:

  <data-role name="ReadWrite">
     <description>Allow Reads and Writes to tables and procedures</description>

     <permission>
         <resource-name>Accounts</resource-name>
         <allow-create>false</allow-create>
         <allow-read>true</allow-read>
         <allow-update>true</allow-update>
     </permission>

     <permission>
         <resource-name>MarketData</resource-name>
         <allow-create>false</allow-create>
         <allow-read>true</allow-read>
         <allow-update>true</allow-update>
     </permission>

       <mapped-role-name>supervisor</mapped-role-name>
 </data-role>

To see how the users and roles were defined for JAAS, see the application-roles.properties and application-users.properties in the EAP_HOME/MODE/configuration/ directory. JAAS uses application-users.properties to determine user credentials and the application-roles.properties file maps the username to the mapped-role-name element.

The Web-Services-as-a-Data-Source quick start demonstrates how to use the Web Service Translator to call a web service that returns an XML document. The quick start also demonstrates how to define a transformation using XMLTABLE and XMLPARSE to translate the XML document into a relational database.

To run the quick start, go to the EAP_HOME/quickstarts/webservices-as-a-datasource directory and open the README.md file in a text editor and read the instructions it contains.

When the quick start is configured, the CustomerRESTWebSvc.war is deployed. This represents the web service data source that JBoss Data Virtualization will access. (It was designed this way so that no external resource is required in order to execute the quick start.) The dynamic VDB, namely webservice-vdb.xml, defines the VIEW using DDL, in order to read and transform the XML document returned from the web service.

The JBoss-Data-Grid-Running-as-a-Data-Source example demonstrates how to access JBoss Data Grid cache using the infinispan-cache translator.

To run the quick start, go to the EAP_HOME/quickstarts/jdg-remote-cache directory and open the README.md file in a text editor and read the instructions it contains.

The source model metadata will be created dynamically based on the configured resource adapter. Notice there is no DDL is being used, as in the local cache quick start. This quick start is defaulting to the dynamically created metadata. However, you can define your own metadata, as in the local cache quick start and provide any necessary overrides.

This example shows you how to do the following:

When you utilize Hibernate, it is normally a one-object-to-one-data-source mapping. By using JBoss Data Virtualization as the data source, you approach the integration at the data layer rather than application layer. This makes it easier to join related information and expose it through hibernate, rather than merge the related data at the application layer.

The Hibernate-on-top-of-teiid quick start demonstrates how hibernate can take advantage of multiple data sources through a single Java object by using the data federation capabilities of JBoss Data Virtualization. This quick start extends the Portfolio VDB and creates a view that is mapped to a single relationally mapped object in hibernate.

To run the quick start, go to the EAP_HOME/quickstarts/hibernate-on-top-of-teiid directory and open the README.md file in a text editor and read the instructions it contains.