Last week, Clouderati Twitter stream was full of back and forth arguments on possible dumping of Amazon API by Openstack. It all started with a recent Silicon Valley Cloud Computing Group meetup on Openstack. One of the takeaways from the meeting is that Openstack is dumping the API compatibility mode. Nebula, the cloud computing project out of NASA, had support for Amazon API, much like Eucalyptus. Once Nebula became part of Openstack, it appears they have taken a decision to drop Amazon API compatibility. This started the twibate among the Clouderati and I thought I will offer my take through a blog post.

So, what is the big deal?
When Nebula adopted AWS API compatibility, the idea is to let users manage Nebula cloud with the same tools they used with AWS, making it interoperable with the largest public cloud provider. There were expectations among some folks that there will be continued support for Amazon API when Openstack.org was launched. Now those people feel let down by this news and starting to wonder if the whole Openstack attempt was a hailmary pass by Rackspace to close the gap with AWS. On the other hand, there are some passionate folks in the Openstack community who feel that the right approach to compete with Amazon is by not following them but by offering an alternative. One of the major reasons for dropping the AWS compatibility is lack of support from Amazon and possible legal risks in the future. Of course, when passions run high in the tech community and the topic contains the term “open”, you will also see doomsday predictions about anything “open”. There are also some folks who say that Rackspace is abusing the term “open”. With VMware wanting to push their cloud offerings to the forefront, there are some spin coming along from the supporters of VMware technology. The net result is a cacophony which, as usual, confuses bystanders and other users wanting to step into the cloud.
My take
When something like this happens with the term “open” in the middle, I get the itch to jump in and offer my soundbites to it. Even though I am an unabashed supporter of anything “open”, I will try to be as objective as possible in analyzing this issue. These are some of my thoughts on the topic.
  • If anyone thinks that the Openstack move by Rackspace is not a hailmary pass, he/she is just being nice naive. Rackspace is a business organization and their move towards Openstack is not altruistic to begin with. This being a hailmary pass by Rackspace is only a very small part of the whole story. By focussing only on this part, some people are missing the big picture
  • It is important for us to realize that Rackspace has put their compute and storage platform under open source license. This, by itself, is big because anyone in this world can set up a Rackspace like infrastructure easily and without spending their resources developing the same thing again. Imagine a hosting company or data center in Europe or Asia being able to set up a cloud infrastructure service to serve the needs of customers in those countries who are required by their governments to keep the data inside their borders. Imagine a small provider setting up a cloud like infrastructure serving a local community where the users want to actually interact with the folks who run the infrastructure. This is the kind of federated ecosystem we are talking about while discussing the potential of Openstack. There are some who claim that such a federated ecosystem cannot scale like the Amazon cloud. They are plain wrong. A federated ecosystem with an open platform can seamlessly scale if the providers figure out a working relationship with other providers using the same platform. I have already written about German provider Scaleup Technologies partnering with Japan’s XSeed Co. Ltd. to allow their customers to their customers to tap into XSeed’s infrastructure and vice versa. They could do it easily from Scaleup’s UI because both the providers run the same Applogic platform. Even though it is not absolutely necessary to use the same platform to burst up in another cloud, I am just arguing that being a smaller player is not a handicap as long as we have an open federated ecosystem. In this sense, Openstack opens up lots of possibilities in the cloud infrastructure world
  • The Openstack design is intentionally componentized. Combine this design decision with a liberal open source licensing. Anyone (including Amazon themselves) can build a component that makes Openstack compatible with AWS API again. The problem here is not Rackspace’s intentions or any misguided direction of Openstack but the actual problem is Amazon’s unwillingness to either open up their API or give a commitment to not sue anyone who offers interoperability with their API. With their margins high, Amazon may not have any motivation to open up their API or play nice with other players in the space. But, if Openstack gains steam like many (including me) expect, Amazon may be forced to “open up”.
  • The other big concern is about Rackspace’s influence on the project. Let us be frank folks. If anyone expect Rackspace to not exert their influence after putting so much money and resources, they are just being naive. Being a for profit business, it is natural for Rackspace to exert their influence on the project. In this age where we want to have our name for buildings in academic institutions after our big donation, expecting Rackspace to just give money and let everyone else (including their competitors) control the project is just short of madness. Being the largest supporter, they will influence the project. What matters is the governance model. When I spoke to Rackspace Openstack folks during OSCON, they assured me that the governance model will be set in such a way that no single company can exert influence over the direction of the project. They also pointed towards 25+ companies, some of them are competitors to Rackspace, joining the ecosystem. Plus folks, Openstack is an open source project with a liberal license. If someone or some organization doesn’t like the governance model, they can easily fork it out and have the right governance model. We all know this about any open source project but we choose to conveniently ignore this in any debate
  • Openstack’s decision is also strategic. If Openstack supports multiple APIs, there is no way they can influence any standardization of APIs. When there are legal risks involved in using Amazon API and a better API is available from Rackspace without the legal risks, it is only natural to support Rackspace API and be an influencer in any API standardization process.
Conclusion
Well, Rackspace’s intentions may have played a role in the dumping of Amazon API compatibility mode from Openstack project but it is irrelevant because Openstack is much more than just a Rackspace hailmary pass. The moment it enters the Open Source space, Rackspace loses control over the code and a strong and vibrant community can easily minimize their impact by pressing for a good governance model. Otherwise, they can just jump like John McCain and say “fork, fork, fork Openstack”. There is nothing we can do about Amazon compatibility. Only good sense from within their organization or market pressure can make them support openness. I wouldn’t worry much about this move and, rather, focus on driving the direction of Openstack project in a direction that empowers its users, in particular, and the world, in general.
Update: Simon Wardley pointed out to me on Twitter that Rick Clark (of Rackspace) has made it clear that they are not dumping AWS compatibility. Plus, unlike a commenter’s inference below, my post is based on the discussions at the meetup. More importantly, even if Openstack is not dumping AWS compatibility, my post still stands. More than anything else, this post is an attempt to highlight how the very open source nature of Openstack empowers the community to take the project in any direction irrespective of what Rackspace intends to do with the project. I hope readers of this post take that message clearly than giving their own inferences.
CloudAve is exclusively sponsored by
Krishnan Subramanian

Krish dons several avatars including entrepreneur in exile, analyst cum researcher, technology evangelist, blogger, ex-physicist, social/political commentator, etc.. My main focus is research and analysis on various high impact topics in the fields of Open Source, Cloud Computing and the interface between them. I also evangelize Open Source and Cloud Computing in various media outlets, blogs and other public forums. I offer strategic advise to both Cloud Computing and Open Source providers and, also, help other companies take advantage of Open Source and Cloud Computing. In my opinion, Open Source commoditized software and Cloud Computing commoditized computing resources. A combination of these two developments offers a strong competitive advantage to companies of all sizes and shapes. Due to various factors, including fear, the adoption of both Open Source and Cloud Computing are relatively slow in the business sector. So, I take it upon myself to clear any confusion in this regard and educate, enrich and advise users/customers to take advantage of the benefits offered by these technologies. I am also a managing partner in two consulting companies based in India. I blog about Open Source topics at http://open.krishworld.com and Cloud Computing related topics at http://www.cloudave.com.

Leave a Reply