Teradata Data Labs and Oracle Data Visualization make a great combo for quickly analyzing new data. When combined, these products enable the user to quickly analyze data and prototype solutions. Both products are self-serve and allow you to upload your data, then drag and drop your way to compelling analysis.
Introduction to Teradata Data Labs
A data lab is a separate dedicated space within a production data warehouse for agile development of new analytic queries that can combine personal, ad hoc, or temporary data with production data. A data lab can be used to quickly analyze, reorganize, model, and prototype a sample of real-time data, and experiment without any risk to production data.
A data lab can benefit your organization in the following ways:
- Model data to see value before it is formally incorporated into the data warehouse
- Access production space quickly and easily
- Manage workloads for data separately from production service level goals
- Explore data in production tables while maintaining data integrity
- Evaluate data on a short-term basis, such as for a merger or acquisition
- Reduce the database administrator workload by allowing users to manage data independently.
Data labs are created and maintained by the Teradata Database Administrator (DBA) or by designated owners depending on the needs of the organization. A best practice is to create a team that defines the needs of the organization and establishes governance, practices, and characteristics for the data labs. The Data Labs portlet allows you to add lab groups. As part of the lab group creation, defaults are defined for each data lab provisioned within the lab group. Additionally, access and approval controls and lab instructions can be established. Lab instructions provide guidance and best practices to help lab users. For example, provide information for loading data into tables using Teradata Data Mover or load utilities such as FastLoad or MultiLoad. Make a lab group private for security or other reasons so the labs in the lab group are only available to a restricted set of Teradata Database users and roles. After the lab group is created, the DBA can provision a data lab within the lab group and can assign lab group owners who approve user and space requests. Some of those requests can be set to be automatically approved. After a request to add a lab is approved, the lab can be immediately used. The DBA can load data from external sources or aggregate data from production tables. If the user decides data is ready to go into production, they can email the owner or send a request. A request sends a message to the lab owner. The lab owner or the DBA can move the tables into production in accordance with corporate policies.
After loading data into a data lab, Oracle Data Visualization is a great tool to use to view the data. Discover the insights hidden in your data, with rich, interactive visuals. Oracle Data Visualization is easy to use, yet powerful enough to perform advanced calculations.
- Automatically visualize data as you drag and drop attributes, chart, and graphs
- Change layouts to present new insights
- Answer questions quickly with online search and guided navigation
- Empower everyone in your organization to uncover the value in the data
This article will walk through the integration of Teradata Data Labs and Oracle Data Visualization Desktop. Oracle Data Visualization Cloud Services would also work well with Teradata Data Labs, but the integration is not described here.
If you already have Teradata, your DBA should be able to set you up with a data lab instance. If you don’t have Teradata and want to experiment with a data lab, you can download a free version. The free version comes preinstalled on a VM. You will need to download VM Player to run the Teradata VM. Using data lab requires the use of a configuration tool called “Viewpoint”. If you don’t have Viewpoint, it is included in the Teradata VM download. There is a full set of installation instructions available with the Teradata download. Below is an abbreviated version.
Installing Teradata Data Labs
The following are the minimum requirements to install Teradata Express:
- Hardware 64-bit processor with VT Extensions
- BIOS Must be configured to support Virtualization
- Space & RAM Minimum 30GB free space and 8GB RAM
Download Teradata Express for VMware with Viewpoint from the below location:
http://downloads.teradata.com/download/database/teradata-express-for-vmware-player
Any recent version of Teradata with Viewpoint will work. For example, Teradata Express 15.0:
Once you are done with the download, unzip the file into a folder. The image was zipped using 7-zip. 7-zip is free and may be downloaded from http://7-zip.org/
If you don’t have a Vmware Player, VMware Player is free and can be downloaded from https://my.vmware.com/web/vmware/downloads.
- Open VMware Player and choose Open a Virtual machine and navigate to the file directory to Teradata Express Folder, looking for the "0_sles10.vmx" file.
VMware will now present you with a dialog box asking if you copied or moved this image. Be sure to choose “I MOVED IT”. The VM will take a couple of minutes to boot (depending upon your host machine size).
- Login into the VM with username “root” and password “root”. The image will take another couple of minutes to initialize.
- Viewpoint is accessed through a web browser. To start the Firefox web browser, in the VM go to left side bottom corner and click on Computer and then double click on Firefox
Viewpoint services are not automatically started. To begin the Viewpoint services click on the "Viewpoint Start" icon on the Teradata Express desktop.
- If you wish to verify the viewpoint status from a Gnome Terminal use the following command:
/opt/teradata/viewpoint/bin/vp-control.sh status
2. Viewpoint is accessed through a web browser. To start the Firefox web browser, in the VM go to left side bottom corner and click on Computer and then double click on Firefox
- Firefox will take a couple of minutes to initialize. The Viewpoint page should come up automatically, but if it doesn't, enter this URL:
Depending upon the configuration Viewpoint may be running on port 9080:
http://localhost:9080
- Login to Viewpoint as username = "admin" and password = "teradata".
3. Once you are logged into Viewpoint, you need to connect to the TD system:
- From the Teradata Viewpoint portal page, click the tools/settings icon in the upper right corner.
Open the Monitored Systems application.
Click the “+” next to “Systems” and select Add Teradata System.
Fill in the requested information. An example screenshot is below:
- Enter a system nickname.
- [Optional] Select the Enable system check box. After it is enabled, the system starts collecting data.
- Enter the TDPID “localhost”.
- For the tdwm user, enter the password “tdwmadmin” (without quotes).
- For the Viewpoint user, enter a login name and password (In the TD16.0 VM the viewpoint user password is “viewpoint” - without quotes). If the viewpoint user doesn’t exist, you will need to create it.
- After you have added the system, close the “monitored systems” application using “close” in the upper right corner.
4. Next the Data Labs application needs to be configured.
- Click “add content” then select the “Data Labs” application
- A “1” will appear in the application area. Click on the “add” button.
- Fill the information to connect to the system.
5. Next, add a Data Lab Group
- A wizard will guide you through the group creation
- Once complete the application will show the different groups and their labs (at this point there are no labs)
- Now that the Lab Group is created, you can deploy a lab. Click on the “+” sign above “labs” and submit the requested info.
6. Add users to the lab.
7. That completes the Data Lab setup. The next step is to load data. Teradata Studio has a Data Lab “view” that makes loading data very easy. The Data Lab View provides access to tables in a Teradata Data Lab. It is a tree view that displays the hierarchy of Data Lab groups with child Data Labs. In the hierarchy, a Data Lab has tables as children. From the view, you can copy, load, and export data into Data Labs. The Data Lab View also allows you to create tables derived from external file data using the Smart Load Wizard.
Integrating and working with Teradata Data labs and Oracle Data Visualization Desktop (DVD)
Oracle Data Visualization gives you the best of both worlds: accurate, consistent analytics you can rely on and the ease-of-use and speed you need to change it. Rich, interactive visuals show the story in your data and make it easy for you to share insights. As easy as you expect—upload your data, then drag and drop your way to compelling analysis.
It is assumed you already have DVD, either directly licensed from Oracle or downloaded from the Oracle DVD download site:
Once installed, DVD needs to be configured for Teradata Data Labs. If you need help installing DVD with Teradata, you can find it here:
Make sure you have configured the “extdriver.paths” file as per the above article. Also, it is important to install “base TTU” as just installing Teradata ODBC drivers is not enough.
1. Add a New Connection in the DVDesktop with Teradata data source.
2. Provide TD connection details and click Save button
3. Now, you should be able to save the connection without any errors and should be able o see the connection listed on Data sources page as below:
4. Now add your data sources to project.
5. To add, select the database and then select required tables to add to the project.
6. Once you select the required table and then click ok, you should be able to see the data without any errors.
7. Now, click on Add Data Source and you should be able to see the newly created data source on the Data Sources page as below:
8. Click on the options list on Tera_labs Data source and then select Create Project.
9. Now, you can see the project recently added in the left side pane.
