Industry News Desk
Cloudera Puts Out New Hadoop
Tested for production use, it’s supposed to have improved performance, greater stability, and extended authentication throughout
By: Maureen O'Gara
Apr. 23, 2011 06:45 PM
Hadoop's prime commercializer, Cloudera, has started pushing its third rev of the popular Big Data open source data management framework inspired by Google's MapReduce as the competition for such widgetry is heating up.
Tested for production use, it's supposed to have improved performance, greater stability, and extended authentication throughout.
It integrates with business intelligence tools and RDBMS systems like Informatica, Jaspersoft, MicroStrategy, Netezza, Talend and Teradata and, besides Apache Hadoop 3, now includes HBase, the Hadoop database for random read/write access; Hive, the Facebook-built SQL-like queries and tables for large datasets; Pig, the Yahoo-developed dataflow language and compiler; Sqoop, Cloudera's MySQL-Hadoop connector; Flume, its own data-loading program; the Hue GUI; and the Zookeeper configuration tool.
Cloudera says the 100% open source CDH3 integrates all components and functions to interoperate through standard APIs, manages required component versions and dependencies and will be patched quarterly.
It supports Red Hat, Centos, SuSE and Ubuntu Linux and can run in the Amazon or Rackspace clouds.
Small MapReduce jobs are supposed to run up to 3x faster and filesystem I/O is up to 20% faster, with 2x improved performance in HBase query throughput.
Twitter is already supposed to be using it broadly.
Cloudera peddles a commercial version with proprietary tools and support.
Reader Feedback: Page 1 of 1
SOA World Latest Stories
Subscribe to the World's Most Powerful Newsletters
Subscribe to Our Rss Feeds & Get Your SYS-CON News Live!
SYS-CON Featured Whitepapers
Most Read This Week