HP Vertica adds analytics tool for semi-structur | steerloaderのブログ

steerloaderのブログ

ブログの説明を入力します。

For the elderly sage from Hebden Bridge, no contemporary politician is quite good enough, and none ever will be, because they will never match the standard set by Margaret Thatcher. For all its limitations, the Tory party was her party, why is why if anyone reading this votes Labour for any reason, it is Sir Bernard Ingham's settled view that you are either irredeemably stupid, or out of your mind.With organizations keeping more data for sophisticated analysis the number of business intelligence platforms keeps expanding.One of the lesser known ones is Hewlett-Packard Co.'s Vertica massively parallel data warehouse platform for analyzing structured data on Linux clusters. HP bought Veritca in 2011.

Now it has announced Vertica 7.0 Enterprise Edition with a number of new capabilities including improvements to the management console and the addition of a Java software development kit.But it also announced a new product called Vertica Flex Zone for exploring semi-structured and unstructured data before importing it into Vertica platform."We think it's going to go well with people who already use Hadoop and want to accelerate the exploration angle," said Luis Maldonado, director of product management, HP Vertica. Other will be satisfied with using Flex Zone by itself, he said.Flex Zone doesn't need schemas to be defined before structured or unstructured data is loaded for exploration. Its connectors support JSON and CSVs comma separated values and other file stores for analytics.

As a result mixed data like application logs and customer information from Web sites can be accessed in several ways. SQL can be run against the data."That we believe is going to be a game changing approach for how you do visualization," Maldonado said.Once interesting data has been found, he added, it can be exported in one-step into columnar form for real-time analytics.Flex Zone "is a great on ramp or discovery tool and a very nice way of making the whole data life cycle — from early storage to early exploration to operational — much smoother," he said.