Open... and Shut The industry's most promising cloud platforms have one thing in common: none of them are winning. At least, not if we equate "winning" with "lots of users" and "lots of revenue" or, more simply put, with "displacing Amazon Web Services."
Even so, competition between the big contenders - OpenStack (Rackspace), CloudFoundry (VMware), Eucalyptus, CloudForms/OpenShift (Red Hat), and CloudStack (Citrix) - is fierce, and for good reason. As VMware vice president of Cloud and Application Services Jerry Chen told me, today's cloud competition is "about defining the platform stack for the next 10 years."
Just ask Microsoft how much it matters to win the platform wars. The stakes are huge.
Some of the most intense competition of late has been between OpenStack, founded by Rackspace and NASA, and CloudStack. Just as Citrix was thumbing its nose at the OpenStack party, leaving for comfortable coexistence with Amazon's APIs and the Apache Software Foundation, OpenStack was getting serious about turning itself into a real, industry-led foundation with Red Hat and IBM finally joining forces.
Sandwiched in between these announcements, VMware celebrated the one-year anniversary of its CloudFoundry PaaS project. The company announced a slew of new partnerships (Cloud9, X.commerce, and others), open-source code called BOSH to simplify release engineering, deployment, and lifecycle management of large-scale distributed services, and growth in its community.
VMware chief technology officer Steve Herrod may have gotten ahead of himself when he said CloudFoundry aspires to be "the Linux of the cloud," but it's a title that each of the major cloud contenders surely aspires to own.
So what's holding them back?
While community around each of the projects has been growing at a healthy clip, only Amazon is knocking the ball out of the park, revenue-wise. Sure, there's lots of tire-kicking. Diego Parrilla, chief executive of StackOps, a contributor to OpenStack, suggests that the momentum behind OpenStack is bigger than the few public references would suggest:
These numbers aren't too far off from what CloudStack, Eucalyptus, and others are seeing. Given all the interest in cloud, why isn't there less tire-kicking and more tire-buying?
The answer may come down to mainstream inability to grapple with web scale infrastructure. As OpenStack's Andrew Shafer puts it:
If an organization isn’t accustomed to building and managing web infrastructure, and is not well versed in practical devops tools and practices, then deploying and operating a significant cloud is going to be overwhelming. That’s true no matter which stack is chosen.
No one is content to blame it on the users and wait 10 years for them to come up to speed, however. The stakes are too big. So the race is on to make deploying cloud applications super easy.
This, of course, is what PaaS is all about. Ironically, PaaS, which almost wholly eliminates the user's need to worry about infrastructure, appears to be a bridge too far. At least for now. Microsoft originally released Azure as a PaaS offering, struggled to get adoption, and has morphed Azure into a credible IaaS product. Why? Because users didn't appear to want to give up that much control. They still wanted to be able to interact with the infrastructure running their apps.
The problem may be that the early adopters interested in IaaS are different from the eventual mainstream majority that will happily embrace PaaS. But we're still in the early innings of cloud, and so IaaS rules.
But perhaps it would do better if deployment were a wee bit easier.
On the CloudStack front, while it's currently focused on successfully transitioning to Apache ownership improving ease of use is also top-of-mind. In fact, Gartner notes that CloudStack already may be better than VMware's vCloud Director for ease of implementation.
This ease-of-use mantra is also top-of-mind for Eucalyptus, CloudFoundry, and Red Hat.
All of which is to say that the cloud platform war is finally settling into a phase that should prove productive for the mainstream production deployments for which the industry has been pining. And despite Microsoft Azure's attempts to compete, the only credible response to Amazon seems to center around openness, however defined. The cloud market is about to get easier to deploy, and hence, a lot more interesting to the early mainstream users that have been mostly locked out of the early phase of experimentation with the various vendors. ®
Matt Asay is senior vice president of business development at Nodeable, offering systems management for managing and analysing cloud-based data. He was formerly SVP of biz dev at HTML5 start-up Strobe and chief operating officer of Ubuntu commercial operation Canonical. With more than a decade spent in open source, Asay served as Alfresco's general manager for the Americas and vice president of business development, and he helped put Novell on its open source track. Asay is an emeritus board member of the Open Source Initiative (OSI). His column, Open...and Shut, appears three times a week on The Register.