Talend, a global open source software leader, today announced that it has added big data profiling for Apache Hadoop to its next-generation integration platform, enabling organizations to discover and understand data stored in Hadoop clusters and to take the guesswork out of finding and identifying problem data. The solution addresses one of the most significant big data problems facing companies today: poor quality data costs U.S. businesses an estimated $700 billion a year due to the inefficiency it causes and the loss of customers, sales and revenue.
Data quality issues are endemic in most organizations. Duplication, incompleteness and inconsistency all play a part in undermining operational efficiency. Data profiling – the process of gauging the character and condition of data stored in varied forms across the enterprise – is a vital step toward gaining control over organizational data. Data profiling is also an emerging, but critical, big data best practice. Without the ability to assess big data and its relevance to their needs, organizations are likely to accumulate data that is at best irrelevant or at worst detrimental to their use of Hadoop and other big data platforms.
“If organizations are not careful, their big data investment can degenerate from a strategic asset into an unmanageable burden,” said Fabrice Bonan, co-founder and chief technical officer, Talend. “Understanding data is essential to using it effectively and improving its quality. To achieve these goals, Talend provides exploratory profiling capabilities to understand the structure and level of quality of data stored in big data platforms such as Hadoop. As the first open source provider of data quality solutions, our ability to summarize details about large data sets from different angles makes Talend a powerful tool for business analysts and data scientists.”
For many businesses, data profiling projects remain stuck in neutral, thwarted by the time demands of building data profiling tools in house, or the steep costs of purchasing a proprietary solution. Talend Platform for Big Data includes a new set of capabilities that provide deep, organization-wide visibility into all big data in its many locations and forms. These include:
- Profiling allows users to analyze their data in their Hive database on Hadoop. Profiling is performed “in place”, meaning data does not need to go through the time-consuming process of being extracted from Hadoop before being profiled.
- Talend data profiling leverages the power of the Hadoop cluster, allowing users to scale up with additional servers to boost performance and deal with increases in volume.
- Analysis provides a custom graphical report on the level of quality of organizations’ data. Data quality analysis includes standard tests that apply to all types of data like empty/missing values, number of duplicates, length of data and shapes of data. It includes further tests for specific data domains such as e-mail validation and phone number validation. These tests can be customized and extended as needed.
 Ovum, "Best Practices for Evaluating Data Quality Tools,” October 2010.
Talend big data profiling will be available in November 2012 as part of the release of Talend v5.2. Additional information and regular updates can be found by following Talend on Twitter at @Talend and on Facebook at www.facebook.com/talend.