Pivotal, the ambitious big data venture led by former VMware CEO Paul Maritz, has finally pulled back the curtain on Pivotal One, a platform for building data-intensive cloud applications. Key elements include Pivotal CF, Pivotal's distribution of the Cloud Foundry PaaS, and Pivotal HD, the company's distribution of Apache Hadoop.
Among its selling points: It makes deployment and management of Hadoop -- normally a pain -- a snap. That's one of the ways Pivotal could stack up against the rest of the PaaS competition as interest in Hadoop as a platform continues to quicken.
Hadoop made painless
The point of Pivotal One is to allow data services to run with the same elasticity and ease as more conventional cloud-deployed applications on top of Cloud Foundry. The data services can also be scaled out, slimmed down, and upgraded to the latest versions without having to monkey with build scripts or tools like Puppet or Chef.
Automation and self-deployment are big selling points for Pivotal, apart from getting Hadoop set up. Applications that bind to Hadoop can have resources within Hadoop (such as HDFS storage) provisioned for them automatically, on-demand. If you have an existing Hadoop deployment, you don't have to tear it down and start over; you can simply hook it up to your new Pivotal CF deployment and pick up where you left off.
Plus, some big names have lined up to contribute to the open source projects underlying Pivotal CF. Verizon Enterprise Solutions is on board as a Cloud Foundry board member and has signed on as a project contributor along with IBM, Intel (itself a Cloud Foundry user), SAP, and many others.
Ease of deployment and management has always been one of Hadoop's weaknesses, which I learned about firsthand when hearing how Oracle has attempted to tame that particular beast. Pivotal One is clearly intended to address that issue and give enterprises a way to build cloud platforms either privately or publicly.
Not the only cloud in town
It would be tempting to call Pivotal the Red Hat Enterprise Linux of PaaS, if it weren't for the fact that Red Hat already has a PaaS product: OpenShift (also available in a private cloud edition, OpenShift Enterprise).
Like Pivotal CF/Cloud Foundry, OpenShift supports a broad range of application languages and has added a good helping of middleware to make it easier for enterprises to make the transition. Red Hat also has a two-year lead over Pivotal, having originally gone public with OpenShift in 2011.
Pivotal might have a few advantages, though. For one, Cloud Foundry's drop-it-in-and-run support for the Spring framework may make it more of a shoo-in for enterprises that already use Spring. (Pivotal is adamant that it wants to be language- and environment-neutral, though.)
Plus, Pivotal's claims of offering a high degree of hands-off automation for the whole process of deploying Hadoop and any related apps is appealing to anyone who's ever pulled out their hair doing just that. By contrast, Red Hat's approach for OpenShift involves Apache Ambari, which provides a UI for Hadoop management but not the kind of spin-up/spin-down integration Pivotal claims to provide.
This story, "VMware spinoff Pivotal launches next-gen cloud dev platform," was originally published at InfoWorld.com. Get the first word on what the important tech news really means with the InfoWorld Tech Watch blog. For the latest developments in business technology news, follow InfoWorld.com on Twitter.