This week I saw two articles that captured the two visions of IT that will dominate the future. Both were interviews with senior IT leaders, one a CIO of a major technology company, the other a senior executive with a leading system integrator. One article depicted a vision of IT as a future of standardized, commodity offerings, while the other portrayed IT as a critical part of every company's business offerings. Two visions of ITs role in stark contrast to one another. Each seems to obviate the other. But is that really true? My take is that both views are true, and the CIO of the future has to push one to make room to achieve the other.
The first article, Cloud Computing: IT as Commodity, is based on an interview with Siki Giunta, vice president of cloud computing and software services at global IT consultancy CSC. The article describes the current Amazon offering as offering computing elements — virtual machines, etc., that require a great deal of resource assembling, system management, security monitoring, and the like. Rather than this approach, Giunta advocates IT leveraging an outside application provider (the article seems to imply it would be a firm like CSC, Giunta's employer) and delivering a service catalog of common applications, e.g., e-mail. With an approach like that, IT manages contracts and user relationships, and constructs a portfolio of offerings from specialist providers who can do a better job at scale than any single IT organization could do on its own. The implication is that IT should get out of the business of running applications on its own and become an integrator of other firms' offerings.
By contrast, the second article, A former "prissy girl" takes on tech, is an interview with Intel's CIO, Diane Bryant. One of her responses to a question portrays a very different perspective on the role of IT:
"I look across Intel's corporate strategy and I can directly tie each one of our pillars to the IT solution that's going to enable it. It's a remarkable time to be in IT. All of a sudden the CIO had better be at the table in the business strategy discussions because they can't launch a strategy without you."
The term I use to describe Bryant's viewpoint is that IT infuses the company's offerings and initiatives. It's not that IT is the offering, it's that the offering can't exist without functionality like capturing transactions and providing analytics, or enabling shared design sessions, or whatever. As I interpret Bryant's remarks, absent IT functionality, Intel can't achieve its business goals. IT is crucial to the future success of the company.
So, which is it? Should IT be focused on streamlining itself and outsourcing its functionality, or should IT be a central component of how the parent company actualizes its business strategy?
The thing is, both perspectives are right, and the Cloud CIO needs to discern which approach is right for each part of his or her application portfolio, and apply the correct IT strategy to fully bring to pass IT's role within the corporation.
There is no doubt that certain sectors of IT have been commoditized and the possibility that an internal IT organization can operate them better than a specialized provider is vanishingly small. An outsourced approach to these sectors is obvious. Tying up capital, operational budget, or precious headcount in supporting a commodity application is clearly wrong. And the first article is quite right in characterizing the use of a cloud provider like Amazon as a foundation for an IT organization running these type of applications as foolish. Managing virtual kit is still managing kit, and managing kit for an application where you provide no significant value or differentiation is a mistake.
That does not mean, however, that a CIO must conclude that his or her job is nothing more than coordinating contracts and putting up a service catalog portal with links to all the outside services. Far from it.
No outside provider of standardized services can possibly help one's company differentiate itself with IT-infused offerings. The only way to infuse a company's business strategy with IT is for the IT organization to implement company-specific functionality. And that requires leveraging IT products and services that are on the leading edge of innovation, because that's where differentiation is possible. And it requires leveraging those products and services in an innovative fashion, taking advantage of the new characteristics they offer — and not treating them as the traditional IT products and services delivered slightly differently.
Let me offer an example. Earlier, I agreed that using Amazon (and this example applies to all cloud providers, so it's not specific to Amazon) to host commodity functionality is a mistake. If you can't provide differentiation in an application area on your own infrastructure, using a cloud provider isn't going to help.
I would go further, though. Using a cloud provider as substitute infrastructure is just as big a mistake. Cloud providers do offer basic computing resources which require significant work to assemble and manage. So the question is, when does it make sense to take on that burden? The answer is, when you can take advantage of the characteristics of the cloud environment to create something difficult or impossible to do within a traditional infrastructure environment. If you're not achieving benefit from using the cloud to do something different, it's not worth the trouble.
This was brought home to me by a tweet response to one I posted about Amazon's announcement this week that it is going to support Oracle's database product. I feel this is pretty intriguing and said so via a tweet. Someone responded "might be interesting, but not until the SLA gap btwn oracle env. and aws is narrowed." In other words, until Amazon provides an environment equivalent to traditional infrastructure, there's no reason to use it.
This misses the point entirely. Cloud providers offer something different than traditional infrastructure and judging their benefit by how well they match what already exists is shortsighted. The reason users are flocking to cloud providers is because they offer something traditional infrastructure can't and assessing cloud computing's benefit should be on the basis of how well it helps one achieve what can't be done in existing infrastructure.
The right approach for a Cloud CIO is to make a ruthless assessment of what the value of each of the applications within the company is. If an application provides no significant competitive advantage, it is a candidate for farming out to a specialist provider. If an application does deliver competitive advantage, it should be an internal IT initiative. The job of a CIO is to execute on the former to provide a funding source for the latter.
Where cloud computing comes into play is the fact that traditional infrastructure is not a very good match for innovation. Slow to provision, difficult to scale, impossible to implement elasticity, it is unsuitable for the kind of efforts Bryant describes. Implementing those kind of applications makes the effort of assembling and managing computing components worthwhile.
Bernard Golden is CEO of consulting firm HyperStratus, which specializes in virtualization, cloud computing and related issues. He is also the author of "Virtualization for Dummies," the best-selling book on virtualization to date.