In a largely downbeat year, I couldn't help but delight in Larry Ellison's over-the-top takedown of cloud computing last September. It was classic Larry: In an onstage interview he asks why, exactly, computers attached to networks have suddenly become "the cloud." "What are you talking about?" he yells over and over.
Fair question -- and not just Ellison is asking it. The twelfth most popular InfoWorld article in 2009 was What cloud computing really means, a 900-word definition written by executive editor Galen Gruman and me 20 months ago. It's difficult to recall any trend about which the basic question, "Now, what is it again?" was still being asked more than two years after it emerged. In our 2008 article we offered a simple answer: Subscription-based or pay-per-use services that, in real time over the Internet, extend IT's capabilities.
[ Get the no-nonsense information you need to take advantage of cloud computing in InfoWorld's 21-page Cloud Computing Deep Dive PDF special report. | Stay up to date on the cloud with our Cloud Computing Report newsletter. ]
While we still stand by that definition, all kinds of issues are mixed up in this fluffy hairball of a trend. David Linthicum has examined many of them this year in his outstanding Cloud Computing blog for InfoWorld. Here, we'll answer five key questions that loomed large on the cloud horizon in 2009.
What defines a cloud service?
In keeping with any trend at the peak of its hype curve, confusion reigns over what is and what is not a cloud service. In our brief 2008 definition, we didn't talk about the definition of a cloud service in detail; we just laid out the main types of services. So based on a growing industry consensus -- and some good work by the analyst firm IDC -- let's try and nail this down:
Self-service: If you can't go to a Web site, set up your account(s), and start provisioning over the Web yourself, then it's not a cloud service. With the self-service approach, epitomized by Amazon EC2, you don't need to spend hours on the phone working out your special configuration. In fact, you can't.
Commodity pricing: Self-service enables cloud service providers to keep costs down. You have an array of pay-per-use (or subscription) options, and if you don't like them, you go somewhere else. If you can negotiate and get some special deal just for you, it's conventional hosting or outsourcing, not a cloud service.
Transparent scalability: If you want more of a service, you go to your service provider's Web site and specify more. Your bill goes up accordingly, but otherwise, you don't have to think about it, whether you're talking about twice the number of VMs or triple the number of seats for a cloud-based development platform.
Shared infrastructure: Cloud services have a one-to-many relationship with their customers. Software-as-a-service providers, for example, deliver their wares in multitenanted fashion, where a single instance of the software runs on the provider's servers and all users log onto that same instance. Infrastructure-as-a-service providers move around customer VMs as needed, so that VMs from multiple customers often run on the same server.
Machine addressability: Just about every major cloud service provider now offers a set of APIs developers can use to call on some aspect of a cloud service, whether a CRM application or a disaster recovery routine. Customers can then build custom capabilities that meet their specific needs on top of the core capabilities delivered by the service provider.
The shared infrastructure aspect is still the main sticking point for potential enterprise customers of cloud services. They don't like the idea of their database tables or server images sitting cheek-by-jowl with those of other customers. Rightly or wrongly, they see it as a security risk.
When services aren't shared, however, and customers get dedicated resources, you have conventional hosting. That may mean up-front costs for hardware or software, the lack of which is one of the chief benefits of the cloud. Even if those costs are spread out, scalability won't be as transparent, and the provider's overhead of maintaining dedicated resources will be passed along to you in cost and longer turnaround times.
Is there such a thing as a private cloud?
According to IDC, the top three objections to cloud services are security, availability, and performance. Not that any of these things will necessarily be worse than what in-house IT has to offer. The problem is that all three are outside IT's control -- which is why, among the enterprise execs IDC surveyed, the so-called "private cloud" was preferred over public cloud services by four to one.
But let's be absolutely clear: The idea that you can transform an entire, existing enterprise datacenter into a centrally managed private cloud, any portion of which could be provisioned for any application on the fly by stakeholders, is nonsense. Public cloud service providers are designed from the ground up for massive scale (and often a specific purpose). Typically, they have fully virtualized infrastructures with homogenous commodity hardware running open source software. Plus, self-service provisioning, multitenanted applications, and chargeback systems do not build themselves.
So we are left with more modest ambitions. We can all agree that at least one technology that underlies public cloud services is taking the data center by storm: server virtualization. But I do not believe, as VMware seems to, that large-scale virtualization and the cloud are pretty much the same. A cloud service, inside or outside the firewall, must behave like a service.
The IBM approach to the private cloud has been quite creative: the cloud appliance. Instead of signing up for a service outside the firewall, you buy a turnkey system that meets most of the criteria of a public cloud service and set it up in-house. IBM CloudBurst appliances are fully virtualized blade servers with self-service Web interfaces and chargeback metering built in. So far, the company has rolled out a CloudBurst appliance for business analytics and a WebSphere CloudBurst appliance for dev and test.
IDC predicts that, next year, IBM will have plenty of company. In its top 10 predictions for 2010, the analyst firm anticipates that "virtually every major systems and application software vendor will introduce 'cloud appliances' as a very simple-to-adopt packaging approach for its private cloud offerings."
But are these clouds or puffs of steam? Put the same functionality as an individual cloud service in a physical box, and you answer objections about ownership and security, but that's about it. Appliances, by definition, are isolated point solutions. If the long-term goal of the private cloud is to move, step by step, toward "cloudlike" manageability of the datacenter and the applications that run in it, then you need to start with a top-down examination of your enterprise architecture.
You'll be glad to know that most of the major consultancies will be happy to help you do that. But it's best to put this broader vision of the private cloud in context -- the "cloudification" of IT is just the latest in a long line of initiatives intended to improve processes and optimize resources over time: reengineering, service-oriented architecture, and so on. The main difference with cloud computing is that, without question, a major effort to deploy server virtualization is a step one.
Will cloud services replace the Microsoft desktop?
Few events in 2009 got more play than the July announcement of Google's Chrome OS, which was billed as a "cloud operating system" and potential competitor to Windows. Or, in less glorified terms, a Linux kernel with the Chrome browser as its command shell. In November the Google revealed that Chrome OS would be the software basis of a Web appliance, and that the applications running on it would be Web apps running in the browser.
In some ways, a Web appliance running browser-based apps seems like a more formidable rival to a PC running Windows and Office than, say, a PC running Ubuntu and OpenOffice. For one thing, an appliance takes off the table the whole hairy hardware compatibility question that has dogged desktop Linux from the start.
Also, the rivalry among cloud-based office suites is heating up. In October, InfoWorld's Neil McAllister wrote a comparative review of three browser-based alternatives, Microsoft Office Web Apps, Google Docs, and Zoho. His conclusion, however, was clear: "When it came down to it, none of the three Web-based productivity suites I tried proved an adequate substitute for traditional desktop software."
But what about the long haul? HTML 5 promises ever richer Web applications -- and Google makes a big deal about Chrome OS supporting HTML 5. For the immediate future, browser-based office suites seem best suited for collaborating on documents rather than replacing good old Office itself. But what Larry Ellison dubbed the "network computer" 13 years ago seems a lot more plausible now than it did then.
Do cloud services mean the end of IT as we know it?
IT tends to scoff at cloud computing. But obviously, there's a fear factor, too. If IT is too caught up in other projects to have the resources to launch some new initiative, the business side may threaten to use a cloud service instead.
The standard line is that cloud services can shoulder nonstrategic chores, like managing e-mail or the company Web site, and free up IT time for more strategic work. As InfoWorld's Leon Erlanger noted in The tech jobs that the cloud will eliminate, the shift is likely to be very gradual, with effects similar to that of other forms of outsourcing. Larger companies are very unlikely, for example, to move their core financials to the cloud in the foreseeable future.
Recently InfoWorld's David Linthicum examined the flip side of this question: the cloud as an engine of job creation. Notes Linthicum: "This seems to be the largest inflection around a hyped space in IT that I've ever seen, especially considering we've been in a downturn in which many companies have reduced IT jobs."
But where will those IT recruits for cloud service providers come from? Perhaps the biggest faux pas of the 2009 cloud season came from Richard Marcello, president of technology, consulting, and integration solutions at Unisys, who noted that the cloud is a great way to save money by eliminating U.S. jobs and hiring workers in India to run cloud services instead.
Does the cloud really enable anything new?
Here's where things get interesting. The reason we chose Google's MapReduce framework as the No. 1 emerging enterprise technology of 2009 is that it enables something completely new: data analysis on a monster scale with commodity hardware.
And as it turns out, the cloud is the perfect place for that kind of one-off, large-scale processing: Upload copies of existing data -- whether a chunk of the data warehouse or gobs of your internal log files -- so you don't need to sweat availability or invest in infrastructure you'll use only once in awhile. No surprise that last April, Amazon added Elastic MapReduce to its roster of cloud services.
And MapReduce isn't the only game in town. Large-scale data processing using columnar databases is an old idea gaining new traction for analytical applications. Standard, SQL-based databases have well-known performance constraints. So why not just upload your SQL data to a specialized processing service in the cloud, have the provider convert it to "NoSQL" format, and run the job in a fraction of the time?
Cloud-based development platforms also provide capabilities difficult to get elsewhere. Salesforce's Force.com already offers a rich environment for building multitenanted Web applications. Microsoft's Azure promises to be packed with component services supplied by Microsoft and third parties, more than you could ever afford to provision yourself.
When people talk about cloud services, they usually focus on low-cost solutions to replace mundane, non-critical stuff IT would rather not deal with, like e-mail or dev and test platforms. But the most exciting potential of the cloud is as a platform for Internet-based services that deliver entirely new capabilities fast without the upfront costs. Businesses that latch on to new services with strategic benefit and integrate them into their existing processes and infrastructure will enjoy a big advantage in the coming decade.