Denodo Platform

Denodo Platform is a data virtualization solution that enables integration, access, and real-time data delivery from disparate on-premises and cloud-based sources.

Before You Begin

It is essential that you have the following installed:

  • Denodo 8.0

  • Java 1.8

Setting Up a Connection to SQreamDB

  1. Under Denodo\DenodoPlatform8.0\lib\extensions\jdbc-drivers, create a directory named sqream.

  2. Download the SQreamDB JDBC Connector .jar file and save it under the newly created sqream directory.

  3. In the Denodo Platform menu, go to File > New > Data Source > JDBC.

    A connection dialog box is displayed.

  4. Under the Configuration tab, select the Connection tab and fill in the data source information:

Field name

Description

Value

Example

Name

The name of the data source

sqream

Database adapter

The database adapter allows Denodo Platform to communicate and interact with SQreamDB

Generic

Driver class path

The path to the location of the JDBC driver required for the connection to the data source

path/to/jdbcdriver/sqream-jdbc-x.x.x

Driver class

The class name of the JDBC driver used to connect to the data source

com.sqream.jdbc.SQDriver

Database URI

The URI that specifies the location and details of the database or data source to be connected

jdbc:Sqream://<host_and_port>/<database_name>;[<optional_parameters>; ...]

Transaction isolation

The level of isolation used to manage concurrent transactions in the database connection, ensuring data consistency and integrity

Database default

Authentication

Authentication method

Use login and password

Login

The SQreamDB role

SqreamRole

Password

The SQreamDB role password

SqreamRolePassword2023

  1. To verify your newly created connection, select the Test connection button.

Note

When adding the JDBC driver in Denodo, it’s important to note that a restart of Denodo may be required. Additionally, in some cases, the SQream driver may not immediately appear in the list of available JDBC drivers. If you encounter this issue, a simple solution is to reboot the machine and attempt the process again.

Limitation

When working with table joins involving columns with identical names and exporting a view as a REST service, the query transformation process can introduce ambiguity due to the indistinguishable column identifiers. This ambiguity may result in unresolved column references during query execution, necessitating thoughtful aliasing or disambiguation strategies to ensure accurate results.