OpenStack turned three years old last week. The project has come a long way in that short time, moving from something that was merely interesting at the time to something that is widely accepted by the cloud technology community today. Indeed, OpenStack-focused conferences are packed. The press loves talking about OpenStack. And the number of investment dollars flowing into OpenStack-based startups is unprecedented.
As GigaOM’s Jordan Novet reported, “To celebrate its own birthday, OpenStack released some numbers showing how complex it’s become and how many people contribute to it. When OpenStack Grizzly hit in April, there were 1,000 contributors and 1.28 million lines of code, up from 20 contributors and 44,000 lines of code at the time of the Austin release. There are 231 companies involved, compared with fewer than 50 at launch.”
While OpenStack, including Rackspace, HP, IBM, and many startups, is clearly the darling of the cloud tech community, the number of installations within traditional IT shops has been lackluster when you consider the expectations that were set. While OpenStack can boast some big names including Best Buy, Bloomberg, Comcast, Fidelity and PayPal, the overall numbers are relatively weak when measured against the growth of their its, Amazon Web Services (AWS).
The OpenStack standard and strategy, generally speaking, are everyone else’s response to the market momentum behind AWS. The project provides an open-source alternative that gets many companies on the same page, in terms of how to build and deploy private and public cloud computing systems. This means you can attack AWS as a collection of technologies, and not just duel it out one-on-one.
While the cloud space is much more complex, the logical segments in the emerging IaaS cloud computing market seem to be:
- AWS (including Eucalyptus for private clouds)
Thus, we’re again asked to place our bets on some standard, which means sets of technologies, versus the de facto standard, which means AWS. It’s helpful, then, to consider the pros and cons of OpenStack.
First, the cons. Brandon Butler over at Network World provided me with Simon Wardley’s thoughts on OpenStack last week which are very interesting. “. . . so today we fin that OpenStack has a significant developer and vendor ecosystem but is [categorized] by a collective prisoner dilemma. Forget fidelity with Amazon there isn’t even fidelity between OpenStack distributions and no well established mechanism of doing this (though first steps have been made).”
In practice, the OpenStack distributions vary a lot in features and functions, as well as intra-standard compatibility. OpenStack is addressing these issues, and I suspect they will evolve in healthy directions over time, if given the time. However, as AWS continues its rapid growth, the window of opportunity for the OpenStack community to get its collective act together is quickly closing.
Indeed, the larger OpenStack players are having trouble getting things off the ground. IBM just lost a $600 million dollar bid to AWS to provide cloud services to the CIA, HP is just getting its cloud computing sea legs, and Rackspace recently missed its numbers. These are not things that denote confidence in those moving to OpenStack-compliant technology.
Moreover, there is the notion of AWS compatibility and interoperability with OpenStack, which is also lacking. There are views, such as that from Randy Bias at CloudScaling, that OpenStack should work and play well with AWS to be successful. However, most in the world of OpenStack, such as Rackspace, view their technology as an alternative to AWS. Thus, AWS compatibility is less important.
To an even greater extreme there are those in the industry who are calling for OpenStack to provide “AWS clones.” The concept of creating “AWS clones” or even providing better AWS compatibility would be a step in the right direction.
The danger is that many would view more investment in working and playing well with AWS as admitting defeat, versus providing an “open source” alternative to AWS. As AWS continues to dominate, those in the OpenStack community will have some soul searching to do, perhaps as early as this Fall
The pros of OpenStack are pretty well known. Standard implementations of IaaS services that span vendors, common APIs and services, application portability, data portability, etc.. I think we get the value of leveraging standards by now, and OpenStack has the same talking points, as did hundreds of other standards that came in the past.
So, should you invest in OpenStack, and thus the underlying technologies? While OpenStack is down, it’s not out. The promise of this technology remains the same, although the implementation could use some improvement. Moreover, the adoption of this technology needs to increase substantially within the traditional enterprise.
The OpenStack community could build something meaningful, if given time. The issue is that they may not have the time, and thus get left behind by AWS, and the thousands that are likely to adopt AWS in the next few years.
A lot really depends upon your own requirements, and that’s what should drive the decisions here. Your best bet is to avoid the religious discussions around AWS versus OpenStack versus CloudStack, etc.. I would work to the technology, and not from the technology, in terms of both strategy and tactical deployments. If you take that approach, you’re likely to find the right path, OpenStack or not.