Amazon has floated a new data-crunching cloud based on Hadoop, the
stuffed elephant open-source phenomenon that attempts to mimic the distributed computing platform driving Google's online infrastructure.
Dubbed Amazon Elastic MapReduce, the new web service is a means of processing vast amounts of data "easily and cost-effectively." It sits Hadoop atop Amazon's existing compute and storage services, EC2 (elastic compute cloud) and S3 (simple storage service).
Amazon has long offered developers the power to run their own Hadoop implementations on its virtual data center. But its new service provides pre-packaged Hadoopiness.
"Some researchers and developers already run Hadoop on Amazon EC2, and many of them have asked for even simpler tools for large-scale data analysis," reads a canned statement from Adam Selipsky, VP of product management and developer relations for Amazon Web Services. "Amazon Elastic MapReduce makes crunching in the cloud much easier as it dramatically reduces the time, effort, complexity and cost of performing data-intensive tasks."
In 2004, Google published a paper describing a software framework called MapReduce that it developed for processing beaucoup data across distributed clusters of low-cost server hardware. Shortly after the paper was published, a developer named Doug Cutting launched an open-source project based on Google's description of MapReduce and its very own Google File System (GFS). He dubbed the project Hadoop, after his son's stuffed elephant.
Cutting's now on the payroll at Yahoo!, where Hadoop underpins at least a portion of Yahoo!'s infrastructure. And the technology is used by countless other online operations, including Facebook and Microsoft's Powerset search engine. A new startup calling itself Cloudera has now commercialized the platform.
Meanwhile, Amazon lets you run the platform on its virtual data center. And now, with Amazon Elastic MapReduce, it offers up immediate access to its own Hadoop implementation. "Using Amazon Elastic MapReduce, you can instantly provision as much or as little capacity as you like to perform data-intensive tasks for distributed applications such as web indexing, data mining, log file analysis, machine learning, financial analysis, scientific simulation, and bioinformatics research," the company says.
Amazon is also offering up sample applications for use on its new Hadoop cloud. As with existing Amazon Web Services, you pay for online compute and storage resources by the hour, as you go. ®