IT purchasing model is flawed?

A couple of days ago, I had a discussion with a CIO of one of the largest manufacturing coampanies in Korea. Among the serveral topics, I like to mention his point on how uncomfortable he is when he's paying for a clustered systems for example.

His argument was like this, if I remember it correctly :

• "It's ok I'm paying for one system
• "Why should I pay for additional system which is usually sitting idle waiting for the primary system breaks down?
  
• "On top of that, I pay for complex interconnect and additonal level of services for this kind of high availability system

It's not that he does not understand IT (he has pretty solid understanding of IT as a matter of fact). As a CIO, he is pointing out the fundamental problem in the cost model for the backup system sitting idle. Our discussion naturally evolved into "software as a service or generic services-delivered-on-the network". Yes! In this new model (shared world), the standby system can be configured to  back up serveral primary systems. So the customer would pay far less. Better yet the  customer does not need to care. The customer pay for the service and quality-of-service, availability simply being one of them.

Like this gentleman, CIOs are asking for more simplicity. Good news is IT is evolving toward that direction. Bad news is the progress is damn slow.

At least Sun is far ahead of competitions when it comes to "simplified" computing services which customers can buy as needed. I mean our Network.Com service. The service as of today is rather limited (to batch style computing oriented work). Over time, it will evolve to service majority of enterprise IT needs.

In the future the computing will be much simpler (to buy and use). How far or close are we? We don't know. One thing is sure: Sun's innovation will make it "close".

Simplicity wins at the end of the day. It's my personal belief. Computing won't be an exception.

Posted at 03:16오후 6월 22, 2007 by myungjong in Personal  |