Beyond the hype of cloud computing
Open cloud is here, but is it really ready?
Despite all of the hype that cloud computing is getting, a recent survey shows that in terms of actual pace of deployment, three popular open source cloud computing platforms are not exactly seeing growth to match that hype.
The results might be surprising for followers of cloud computing, but not to the company that conducted the annual "State of the Open Source Cloud" survey, Zenoss. This news is just confirming what they already knew: that cloud is a ways off from being a ready-to-use tool and customers know it.
Just ask Floyd Strimling, VP of Community at Zenoss. Actually, you probably won't have to ask him - he'll most likely tell you anyway. From Strimling's point of view, cloud adoption is very much like the perceived U.S. political landscape, with a very loud one percent actually adopting and hyping the cloud, with the rest, "the 99-percenters," having little to no idea what to do with cloud computing.
For Strimling, the juxtaposition of the survey results reflects exactly that. When 50.5 percent of respondents state that they are planning to deploy OpenStack (with 18.3 percent looking at CloudStack and 9.2 percent at Eucalyptus), that sounds like the OpenStack marketing machine is doing quite well. But then when asked about the timeframe of that deployment, only one quarter of the survey respondents (25.9 percent) were planning to do so in 2013. All of the rest were at least one or two years out from deployment, or more.
This is purely a matter of maturity, Strimling explained. When you look at OpenStack, you see a collection of separate products, like Swift, Nova, and Quantum that have been gathered together and re-branded under one brand.
"That's both OpenStack's biggest strength and weakness," Strimling said. There are a lot of great tools under the OpenStack roof, but they are as yet not organized within one single product. There's very little for potential OpenStack customers to download and actually use or even try. This is opposed to CloudStack and Eucalyptus, which have installable products ready to go.
Right now, the OpenStack community is in the middle of a race.
"Who is going to commercialize it first?," Strimling pondered.
The first company that can put together a solid package of OpenStack offerings will very likely be the one that makes the best go out of using OpenStack technology. With a solid product offering to tout, that vendor will have a better shot to compete with CloudStack and Eucalyptus, who are ahead in this race.
Currently, Citrix' CloudStack and Eucalyptus are on the clear path towards commercialization, Strimling emphasized, since both vendors have a product that is downloadable and basically configurable (though still on the complex side).
Eucalyptus, he added, is the potential disrupter in the race.
"Their relationship with a true dominant public cloud provider is the wildcard," Strimling said. With its ties to Amazon Web Services, Eucalyptus could get a big jump on the rest of the open cloud community.
Who will win the race is anyone's guess. Just figuring out who will be the first in the commercialization of OpenStack mini-race is tricky enough. Strimling would have laid odds on Piston Cloud, which seemed poised to get something together, but now he's looking hard at Linux vendor Red Hat, which is putting a lot of effort into getting a package together. A big focus of SUSE's discussions at its recent SUSEcon conference were also about productizing OpenStack using Dell's Crowbar technology.
But until there's an offering readily available that customers can actually get and use, cloud computing isn't going anywhere fast.