Vendor View

Plenty of Room for both OpenStack and CloudStack

With competing cloud stacks battling it out publicly, customers will be the eventual winners.

• By Shawn Edmondson
• 05/02/2012

The corporate line is that it was a move born of necessity with OpenStack seemingly stuck in a state of awkward adolescence. But we all know that this is probably only part of the story.

It's a move that seeks to upset the balance of power in the cloud wars by performing something of a judo flip on OpenStack, capitalizing on the considerable momentum it has generated and satisfying pent up demand for an enterprise-grade open cloud platform.

It's a bold move and one that I believe will pay off for Citrix. But is it game over for OpenStack?

Not a chance. If you ask me, this is game on!

We have a tendency to treat competition as a zero-sum affair: If one wins, the other must lose. But this is simplistic, particularly in markets as vast as cloud computing. Cloud is one of the most significant architectural shifts in the history of IT. It will fundamentally transform how we purchase, consume, provision and scale IT services. It will yield untold winners -- and losers, for that matter.

There is plenty of room for both OpenStack and CloudStack to survive and thrive.

It's easy to lose sight of the fact that competition is what drives innovation and quality. Competition makes markets. In the arms race for cloud, there's room for both open and commercial options.

Open clouds like OpenStack and CloudStack drive innovation and put downward pressure on commercial pricing. Likewise, commercial clouds provide the assurances and indemnifications that many enterprises desire or require for mission-critical usage.

But there's also plenty of room for multiple open source clouds.

BSD is a stronger operating system thanks to Linux, which forced the community to innovate and specialize. Likewise, SUSE and Canonical have made Red Hat stronger as an innovator and a highly successful publically traded company. Competition is the invisible hand of free markets, which separates the winners and losers by the measures of quality, innovation and specialization.

As OpenStack and CloudStack lock horns in the battle to dominate the open cloud, we should expect something well beyond the standard feature-for-feature horserace for market dominance.

Instead of convergence and competition in a race to the bottom, we should expect divergence and competition through differentiated specializations. Linux distributions are a useful analog.

With the exception of overt clones like CentOS, the Linux battlefield has been defined by specialization. Today, there are scores of specialized Linux distributions spanning industry-, language- and use case-specific variants that range from distributions that apply to large market segments to those that target the most specialized long tail segments.

This sort of specialization is the mark of a healthy and vibrant ecosystem. We should encourage the same for cloud.

How exactly this will play out is hard to know. Perhaps the first split will occur along enterprise and service provider lines, with the preponderance of service providers opting for CloudStack, which doesn't have the complication of a patron of service provider origin. Then again, CloudStack strongly supports the Amazon Web Services (AWS) APIs, which cuts both ways for MSPs. On the one hand, AWS compatibility means easier migration of AWS workloads onto a cloud platform. On the other hand, it also means workloads can move the other way.

But the point is that, just as we shouldn't expect a one-size-fits-all cloud, we shouldn't expect a single winner in the battle for open cloud dominance. Ultimately, two active and vibrant open source communities will drive us toward standardization in a way that selects the best attributes from each without creating an unfair advantage for any specific commercial vendor.

Today, the de facto cloud standard is almost unquestionably AWS, which is an impressive offering with an innovative architecture. But that's not to say that AWS should be the enduring standard -- that creates an imbalance of power in the market that is good only for Amazon. It may be the closest thing we have to a cloud standard today, but we should encourage competition as the basis for evolving the standards that promote rather than constrain a healthy ecosystem comprising, not one winner, but a number of winners that raises the tide for all ships.

After all, this cloud thing isn't a zero-sum proposition. It's a massive transformation with room for many winners. Let the games begin!