Memo to Microsoft: don't bet against Amazon
Private clouds will roll over
Open...and Shut The technology world used to be fairly easy to understand. For a time, IBM dominated the back office while Microsoft monopolized the desktop. More recently, Microsoft and Linux split the difference on servers while Oracle bought the known universe to dominate enterprise middleware and applications.
But along comes the cloud, and everything is up for grabs. Everything.
The incumbent vendors want to pretend that the cloud is business as usual. Red Hat makes it easy to run instances of RHEL on Amazon, while Microsoft's Windows Azure tries to extend its server and database lead to the cloud. Oracle's cloud vision is much the same, except with Oracle products playing center stage. But these and other incumbent vendors may end up disappointed.
The reason is that the cloud creates all sorts of scaling problems, problems that require novel solutions. Fast ip co-founder Benjamin Black suggests that users will bury their preconceived notions of ideal technologies or vendors to achieve scale:
Lost in all the debates about SQL vs. NoSQL, ACID vs. BASE, CAP, and all the rest is simply this: focus on the process to build a great company and great products. That's what every successful technology company has done in order to reach the scale where things like BigTable and FlockDB are required. They didn't become successful because they built these systems. They built these systems because they became successful. If your success drives you to the scale where relational databases are no longer efficient, don't be afraid to look beyond them. That might mean adopting one of the existing NoSQL systems, or even building your own to meet your exact needs. This is what happens at scale.
This quest for and realization of scale initially has companies stopping off at private cloud computing, which favors incumbent technology giants, but it's fast leading to a more disruptive embrace of public cloud providers, according to Barron's Mark Veverka.
The first step away from old-school vendors is new-school private cloud providers like Rackspace's OpenStack, Eucalyptus, Cloud.com, and more. Importantly, not only are these companies new, but they're using the underdogs' preferred tool, open source, to unseat calcified purchasing bureaucracies. 'Try before you buy' effectively translates to 'try me rather than buy them.' And a rising number of organizations are buying into cloud computing according to Gartner.
This wouldn't matter if the public cloud were to remain unpalatable for large enterprises fearful of security and performance issues sometimes associated with public cloud computing. But it won't. Already Gartner is projecting 100-per cent adoption of public cloud computing by Global 2000 organizations by 2016. One hundred per cent means "all of them." The public cloud threatens to be a public nuisance for incumbent technology vendors.
Why? Because, as Redmonk analyst Stephen O'Grady argues, we may see traditional incumbent vendors outmaneuvered by the scale public cloud providers have mastered. The benefits of cloud computing can be achieved in a limited way with private clouds, but are likely to only be fully realized in the public cloud, and the profits from the public cloud will almost certainly congregate in the new kids on the enterprise computing block like Amazon and Google:
Besides the economies of scale that Amazon, Google and others may bring to bear, which might be matched by larger commercial institutions, the experience of web native entities operating infrastructure at scale is likely to prove differentiating. It's not just a matter of running a datacenter at scale; it's knowing how to run applications at scale, which means understanding which applications can run at scale and which cannot. The cloud "public or private" must be more than a large virtualized infrastructure. Whether or not private cloud suppliers and their customers realize that yet is debatable; whether Amazon, Google et al do is not.
Along the way, we're seeing serious grappling for the hearts and minds of developers in the public cloud, most recently in the tug-of-war between Rackspace and Amazon. Some data favors Rackspace, while other data favors Amazon. And while both suit different needs, there is one common driver for both companies' success: the need for scale.
Beyond scale, however, Amazon has another thing going for it. Two things, actually. The first is developers. Amazon has been very effective in reaching out to developers and making EC2 a haven for developer experimentation. Rackspace has also done well with developers but its announced strategy of moving into the "managed cloud" and offering more expensive, premium services threatens to undermine its competitiveness with developers.
Which leaves Amazon to duke it out with Google, another main contender in the public cloud. In its fight with Google, however, Amazon's second big advantage comes into play: focus. Unlike Google, which dabbles in just about everything, Amazon is very focused, as Chris Dixon articulates.
The search for scale will significantly disrupt the technology industry, including enterprise software. Given the importance of cloud computing in this, and particularly the public cloud, it would be tough to bet against Amazon, the company that knows how to scale cloud applications better than anyone. ®
Matt Asay is chief operating officer of Ubuntu commercial operation Canonical. With more than a decade spent in open source, Asay served as Alfreso's general manager for the Americas and vice president of business development, and he helped put Novell on its open-source track. Asay is an emeritus board member of the Open Source Initiative (OSI). His column, Open...and Shut, appears every Friday on The Register.
Not really cloud hype...
I guess it looks a bit like Amazon cheerleading, but that's not the intent. The intent is to look at what scale is doing to the industry, and (at least initially) it clearly favors those, like Amazon and Rackspace, who understand how to scale. Google, too. I don't count out Microsoft (despite the title given to the piece), but I think it has a long way to go to learn how to think beyond isolated computing units (desktops, servers) and instead grok the potential power behind Azure, rather than just the basic operations.
At any rate, I also should have talked more about Rackspace, which now has a significant asset - OpenStack - that could be of significant benefit in its counterargument to Amazon. But I had to stop writing at some point. :-)
Not very meaningful
Your "entire infrastructure" being ... what? A multi-billion dollar banking concern or a couple of guys in a coffee shop? If you can't describe what you're talking about, then you shouldn't be writing.
Oh leave it out
Our entire infrastructure has now been running on Amazon Web Services for several months and is doing very well for it.
If you can't work your way beyond 1990s technology then you shouldn't be working in computing.