First and foremost, I want to be very clear. In no way is this any sort of indictment of OpenStack from a technology or community perspective. Clearly the technology has value, and the community has shown an ability to organize, govern, and rapidly evolve a large platform over the past few years. This should not be overlooked nor understated for a project of this scale. With that behind us, let’s look purely at the macro market opportunity for OpenStack, and more importantly, the vendors trying to monetize it.
Perception Versus Reality for OpenStack
If you listen to many, you would think that OpenStack has or will very soon dominate the world. Pundits will point to the year over year growth of summit attendees, major companies with OpenStack projects in flight, and ever expanding developer and vendor support base. No doubt these are important elements for the long term viability of the technology. Many even consider it a “movement”.
In many ways, however, there is a “if we build it, they will come” attitude about the project, with relatively little revenue to show for it. You can argue the project is in its early stages, but as the OpenStack community itself will remind you, the speed of its evolution has been unprecedented. 451 Research claims OpenStack-related revenue will exceed $1 billion by 2015. While impressive, let’s look closely at the numbers. Most of the revenue is services, with product sales a distant second – $680m and $82m by end of 2014, respectively. Predicted growth rates are impressive, but compared to VMware’s revenue alone, at over $5b for 2013 and growing fast, it’s a small piece of the pie for the 200+ companies sharing the $1b in sales for OpenStack-related products and services. Why compare to VMware? Because VMware’s Software Defined Data Center is what OpenStack needs to unseat in order to grow revenue in existing Enterprise accounts. Greenfield opportunities of course are a different matter, but then again, public cloud is coming…
OpenStack Versus the Cloud
Wait.. isn’t OpenStack actually cloud? Well, no, it’s infrastructure that can be used to power clouds, private or public. It’s basically a set of tools that you can install to create a cloud. So if we think about “the cloud”, as we hear about in popular lingo, what’s generally meant is public cloud services. Obviously there is a software infrastructure component to this, but most consumers neither see nor care about it. There’s good reason why OpenStack’s growth opportunity is mainly limited to the private cloud. There are actually 3 good reasons, 2 of which are giants, and one very famous up and comer. To spell it out, there’s Amazon Web Services (AWS), Google, and Microsoft. AWS is the overwhelming market leader by far, not only providing infrastructure directly to consumers, but also to other companies which in turn deliver “higher stack” services to their customers. And its influence and scale grows seemingly by the minute. Google of course runs most of the web applications we consume as end users on a daily basis, and is now expanding into head to head competition with Amazon with its GCE. And let’s not forget Microsoft, the juggernaut who now sees cloud (public and private) as its ticket to long term viability. While vastly different in approaches, the 3 players have one thing in common: none of them offer OpenStack based services. Given their entrenched positions and existing investment, it’s unlikely they will. So any real challenger or challengers who intend to offer OpenStack based services face not 1, but 3 behemoths to compete against. With AWS revenue estimates alone at over $5b for 2014 and growing at least 30% per year for the foreseeable future, it’s hard to imagine unseating that any time soon.
And let’s not forget, at the end of the day, the public cloud consumer (be it individual or business), doesn’t care what the infrastructure underneath is, as long as it works and provides value. They are certainly not interested in “movements” nor the communities behind them, as this is all completely abstracted from their services. Sure, you will always have conscious buyers who refuse to do business with companies who don’t use technologies they like, but this is not the same as boycotting clothing manufacturers for using slave labor in developing countries. No children are being exploited – it’s simply a matter of preference, and the amount of people who actually care are in an extreme minority. Quality of service and business continuity are far greater concerns to most public cloud consumers.
Why is public cloud so important? Simple: its growth rates are astronomical, with megatrends such as Big Data driving it. And you can bet that much of the spending is from businesses and enterprises, likely taking away greenfield “home grown” infrastructure opportunities where OpenStack has much of its potential.
In summary, OpenStack is and will continue to be successful in offering an alternative platform for cloud and data center infrastructure. In some cases it may even prove superior to other offerings. But from a mass market perspective, the tooth and nail fight from established enterprise vendors on one end, and the turbocharged growth from massive public cloud providers on the other will mean that it will remain in a niche position for some time to come.