As widely reported, Microsoft has begun the process of unveiling details about the pricing and services for its much-anticipated Azure cloud platform. Mary Jo Foley has a pretty complete round-up of the news here. There’s a bit of a debate around whether Azure or Amazon will be the cheaper cloud, but the issue of price isn’t the main point of comparison. The real question is which company plans to support real enterprise apps vs. much less complex and, frankly, more readily consumable, consumer apps.
My vote for the latter is Amazon, but if you want to run real enterprise apps, with lots of appeal to existing enterprise users, I’m betting Azure will be the cloud to choose.
Don’t get me wrong, I love Amazon, and I think in general they are the best consumer commercial site on the Internet. But I’m not sure that Amazon.com is the proper model for an enterprise-class cloud, and, as such, I worry that there’s not enough enterprise-class DNA at Amazon to get the enterprise cloud right.
Microsoft, on the other hand, actually does get the enterprise. After all, they own some pretty enterprisey applications — Dynamics AX, NAV, and GP are world class — and they have this tech stack that is running major corporations all over the world. Moreover, the fact that Dynamics, however small its contribution to Microsoft’s humongous revenue stream may be, has a lot of important enterprise experience and knowledge has been more widely acknowledged in the race to bring Azure to market than many realize. That infusion of enterprise IP has had a role in shaping Azure’s design specs and future, and, from what I’ve been told, the Azure folks have been listening closely. The result, as Azure matures over the next couple of years, will be a seriously enterprise-class offering, with a host of native enterprise-class services, that are accessible to the many many developers schooled in the Microsoft development world.
And, just to make sure the notion is clear: Microsoft will be adapting its own services to its own cloud, which promises better integration and faster time to market than would be the case if the ancillary enterprise services came from third parties exclusively.
Meanwhile, over at Amazon, there is no such internal enterprise product set to morph over to the cloud. There is the one of the world’s best user experiences to draw on, one that has been touted as the archetype for many SaaS-based services. The point is well-taken, most recently by Dan Woods in Forbes, but Amazon’s archetype for SaaS isn’t really enterprise-class, and the SaaS apps that Woods cites as drawing competitive advantage from being based on an Amazon archetype don’t begin to offer the breadth of coverage that is needed in a full-bore enterprise app. It’s easy to make e-commerce, and salesforce automation look as easy as Amazon does, it’s an entirely different matter to present complex, enterprise-wide business processes in a simple, easy-to-use, consumer-like user experience. The reason it hasn’t really been done, IMO, is that it’s largely impossible. Buying cool stuff on a cool website is a lot different than running an enterprise process. Amazon is arguably the best at the former, but it has no experience in delivering the latter in its cloud.
So, Azure is coming, and comparisons to Amazon are a little off base. If I were Microsoft, I’d be teaming up with Amazon and worrying about the real competition: IBM and Oracle. And if I were Amazon, I’d be looking over my shoulder at Google, not Microsoft. IBM has its own cloud initiative well under way, and if I’m not mistaken, once Sun is swallowed by Oracle, we’ll start seeing a pretty comprehensive cloud offering from Redwood Shores as well. These two players will clearly be competitors in the enterprise cloud, and both would gladly take market share from Azure if possible. Truth is, cloud demand will be high enough to avoid such a zero-sum competition, at least from the get-go. And Google, also not a threat in the enterprise cloud (despite the opinion of the likes of Dion Hinchliffe), does get the consumer world, and will be a major headache for Amazon going forward.
The bottom line for Azure vs. Amazon is that it’s the wrong battle to bet on. All clouds will not be created equal, as is the case with all forms of infrastructure. Some roads are for racing, some roads are for moving cars through crowded urban areas. Each is designed and used differently. It’s the same with clouds, and that’s a good thing. Amazon and Azure will both do well, just not at each other’s expense.