Teradata brings DevOps into the data warehouse with Python module

A Teradata module for Python promises IT pros the ability to develop DevOps-fuelled applications.

Teradata: Application aims to make it easier for programmers to use Python in Big Data applicatopms Photo: Courtesy Compact Solutions

Teradata has launched a module for Python which it says could make it easier for programmers and data scientists using the language to create applications that use data in the Teradata Database, and which also promises smoother data warehouse operations.

Python applications can run on an application server and send SQL queries to the Teradata Database, or run within the Teradata Database; programmers can use all the capabilities in Python libraries for advanced analytics or data manipulation. According to Teradata, publicly available Python libraries include the Python Standard Library, NumPy/SciPy, Biopython, Pandas, Mlpy, and Dateutil/Pytz.

As Teradata pointed out, while organisations will already have multitudes of different applications running millions of queries but "the challenge is that applications are not static; they must constantly evolve to meet the ever-changing needs of the business".

Teradata believes that through the emergence of DevOps there is now a bridge between software developers and data warehouse operations which will, hopefully, make it easier for both groups to create and continuously upgrade and manage applications.

According to the company it developed the Teradata Module for Python, "by leveraging the DevOps practices learned from its own product development and the most successful data warehouse users around the world".

This means that when developing Python applications organisations should not have to, "recreate coding standards and tools for consistent operational logging to enable automated monitoring", Teradata said in a statement.

The facilities offered in Teradata for Python include:

Consistent application tooling and logging: The Python module reduces the need for hand coding, according to Teradata and offers consistent activity logging and impact analysis capabilities.

Easier connection to Teradata Database: The Python applications connect to the Teradata Database through Representational State Transfer (REST) services from, "any device, anytime and anywhere or standard ODBC (Open Database Connectivity) drivers", says Teradata.

Application execution in addition to query execution: In order to support administrators overseeing operations, applications built in Python can capture a script version, run ID and execution time for version impact analysis and analysing applications as well as not queries, the company says.

Python Database API Specification v2.0: Implements the standard Python interface to databases.

According to the president of Teradata Labs, Oliver Ratzesberger, Teradata for Python will, "bring DevOps practices to the data warehouse environment" and the Python module will, "allows customers to easily build DevOps-enabled applications, which provide version control, configuration management and the logging of activity".

The Teradata Module for Python package is now available and can be installed directly from PyPI. The open source code is released to GitHub and the documents are available on the Teredata Development Exchange community site for the Teradata Database.

Further Reading:

Facebook and Teradata on Apache Presto and the disruption of open source

Combining analytics and security to treat vulnerabilities like ants

Teradata rolls out platform updates designed to connect Hadoop ecosystem