Need for 'Big Data' Analytics Drives Vendors' Acquisitions
Acquisitions such as Teradata's planned purchase of Aster Data Systems are driven by the growing demand for technologies that can help enterprises mine massive volumes of unstructured data, analysts said.
Fri, March 04, 2011
Computerworld — Acquisitions such as Teradata's planned purchase of Aster Data Systems announced Thursday are driven by the growing demand for technologies that can help enterprises mine massive volumes of unstructured data, analysts said.
Teradata, which already has an 11% stake in Aster Data, said it would pay $263 million to buy the remaining 89% of the data warehousing startup. The deal is expected to close some time in the second quarter.
The acquisition will give Teradata new, massively parallel, in-database analytics and graph analysis tools for addressing emerging data mining and business intelligence applications for what's known as "big data."
The Aster acquisition continues a string of similar announcements from other vendors recently. Examples include Hewlett-Packard's (HPQ) purchase of Vertica last month , IBM 's acquisition of Netezza in September and EMC 's purchase of GreenPlum last July.
"What this says is that the vendors realize they need to really embrace all of the new data types" that enterprises want to mine and to analyze these days, said James Kobielus, an analyst with Forrester Research (FORR).
A growing numbers of companies are trying to see how they can take advantage of unstructured content such as clickstream data, weblogs, social media content and even Twitter streams to deliver new sales, marketing and customer support applications, Kobielus said.
Traditional relational database management technologies have been unable to handle such unstructured data . There's been growing interest in technologies such as Aster's that are optimized for handling large volumes of both structured and unstructured data, Kobielus said.
Aster Data's technology is based on an emerging standard for advanced analytics called MapReduce that leverages massively parallel architectures for processing terabytes and even petabytes of data at much faster speeds than traditional data warehouse technologies.
Aster's Data-Analytic Server technology is a massively parallel platform that is optimized to store very large volumes of data and to perform in-database processing and analysis to deliver faster performance than traditional RDBM systems.
"If you look at each of the vendors that have been acquired, they had figured out ways to improve on what was being done by the established players," said David Menninger, an analyst with Ventana Research.
To compete, established vendors are being forced to either build or buy these technologies, he said. "Buying allows them to get into the market more quickly," Menninger said.
In Teradata's case, some of its existing technology overlaps with what Aster will bring to the table, Menninger said. Even so, Aster's expertise with unstructured data appears to have been "attractive enough to warrant making the purchase," he said.