Cloudera hangs (more) elephants in sky

Big Data in the Amazon-free cloud

By Cade Metz in New York • Get more from this author

Posted in Servers, 2nd October 2009 16:49 GMT

Free whitepaper – Avoiding costs from oversizing data center and network room infrastructure

Hadoop World Cloudera - the commercial Hadoop outfit - has launched beta programs for running its stuffed elephant distro on the sky-high compute services run by Rackspace and SoftLayer.

Based on research papers describing Google’s proprietary software infrastructure, the open source Hadoop platform is a means of crunching epic amounts of data across a network of distributed machines. Named for the yellow stuffed elephant treasured by the son of project founder Doug Cutting, it now underpins online services operated by everyone from Yahoo! and Facebook to, yes, Microsoft.

Cloudera - which Cutting now calls home - already offers versions of its distro for Amazon's so-called infrastructure cloud, EC2, and VMWare's imminent vCloud. And this morning at the Hadoop World developer conference in midtown Manhattan, Cloudera's Christophe Bisciglia introduced beta offerings for running the Cloudera distro on the online infrastructure services from Rackspace and SoftLayer.

Bisciglia tells The Reg that the company has demonstrated management scripts that deploy the Cloudera distro to the Rackspace Servers service and that it has deployed a bare-metal version of the distro straight to SoftLayer's CloudLayer service. Unlike Amazon or Rackspace, CloudLayer's remote service offers direct access to physical machines.

Cloudera product veep Jeff Hammerbacher tells us that most Cloudera customers still prefer to run the platform inside their own internal data centers, but the common assumption is that the future lies high in the sky. ®

Free whitepaper – Standardization and Modularity in Network-Critical Physical Infrastructure