August 7, 2012
The last few weeks have been interesting around the OpenStack ecosystem. We’ve had HP moving object storage and Cloud CDN to general availability. We had Morphlabs introduce an interesting combined hardware and software offering called mCloud Helix. The product is powered by OpenStack, and combines that with SSD-powered nodes to deliver a compact rack mount unit that enabled SMBs to leverage their own private cloud.
Around the same time Dell came to the party announcing that their OpenStack-Powered Cloud Solution is now available with additional server, networking and services options. DreamHost is planning a build out an OpenStack offering. HP announced the general availability of some of its OpenStack based product. While it has long been mentioned that signing a partnership doesn’t make an ecosystem – the release of real products from multiple vendors certainly does.
At the same time we had the announcement at OSCON that the OpenStack foundation was opening up membership to individuals and an indication that the foundation is finally becoming a fully independent entity. Yes some people have some lingering concerns about the finer details of foundation structure, but indications are that it’s going the right way.
Personally I don’t have any bias in the open cloud situation – while Rackspace does indeed support the creation of the CloudU curriculum, CloudU has always been vendor neutral and I’ve never held back from giving my honest opinion. Quite simply I’ve watched over the last two years since OpenStack was first launched and seen something that started as a fairly nebulous grouping of different companies, and has become something real.
At OSCON recently, I joined Alex Williams of TechCrunch and Krishnan Subramanian from Cloudave to discuss the future of the cloud. We spent quite some time talking about Krish’s “Federated Clouds” notion, and the important part that open source can play in that. I tended to push back against Krish’s positive, but largely utopian ideal. The fact is that history has shown that commercial imperatives win, and there will always be some kind of block to completely free federability. That said however, it seems to me that the best way to counter these barriers is to remove as many of them as possible. While some would argue that an open API is enough to do so, I’d agree with Krish that the more that is open the better, and conceptually an open cloud operating standard does more to ease this portability than do open APIs.
It’s early days in this journey, and frankly I don’t see this as any kind of a war. I’m not dogmatic at all about open source, nor am I dogmatic about the battle between open and closed APIs. What I am passionate about is making sure that organizations can move to the cloud, secure in the knowledge that they have choice and flexibility. From a purely conceptual perspective, it seems that the more open a technology is, the more likely that is to be able to occur.
OpenStack isn’t a done deal – but if the past couple of years are anything to go buy, this horse looks set to run a good race.