There's a lot of blogosphere chatter these days about "private cloud" computing. Unfortunately, there's no such thing.I spent the day visiting with the IT team at Fiserv, which provides back-end services to financial institutions. CIO Richard Jones and his team are in the midst of a massive consolidation project across Fiserv's 55 business units.

They spoke at length about the need to standardize functions and eliminate duplication across those units, as well as consolidating a couple of dozen data centers into a few. After a while I realized that what they were talking about was essentially utility computing.

"We want our businesses to be able to say 'We need a certain amount of capacity and performance,'" said Kevin McDearis, Fiserv senior VP for engineering. "We'll worry whether it's on a physical server, it's virtualized, or on the mainframe -- so they can focus their business on providing services to their customers."

That's a pretty good description of the advantages of Amazon's EC2 cloud-computing offering, except it's proprietary and runs on the company's own machines. That's a crucial distinction.

Guest blogger George Crump maintains there's no distinction: "Cloud computing and cloud storage isn't limited to something that uses the public Internet to make connections."

I disagree. As my colleague John Foley points out, "private clouds" is "an oxymoron since cloud computing, by definition, happens outside of the corporate data center." John adds, "it's the technology that's important here, not the semantics."

I actually think this is a case where semantics matter. It's important not just to vendors but to potential users to be clear on the differences, and benefits, of cloud computing via the public Internet vs. utility computing within the enterprise. Both have their advantages and disadvantages (outlined well in Crump's post and in this post by Ken Oestreich on the Fountainhead blog), but this nascent field will be dogged by uncertainty if people confuse the notion of cloud computing with that of in-house utility computing.

For more insight into different forms of utility computing, have a look at Eucalyptus -- Elastic Utility Computing Architecture for Linking Your Programs To Useful Systems, which is "an open-source software infrastructure for implementing Elastic/Utility/Cloud computing using computing clusters and/or workstation farms." I'll write more on Eucalyptus in a future post.