News Stay informed about the latest enterprise technology news and product updates.

Red Hat: Debating enterprise Linux momentum futile

The days of defending Linux momentum in the enterprise may rapidly be coming to a close. Red Hat executive Mark de Visser gets to the heart of the matter with, pointing out that moving from a proprietary Unix system to Linux on Intel results in "10x price-performance improvements" in many cases. De Visser covers more in this question-and-answer interview, specifically the inhibitors to Linux adoption in the enterprise, Red Hat's focus on certain vertical markets, customer demands and the maturation of the open-source platform.

What are you hearing from customers about the enterprise-readiness of Linux and what place Linux holds on a company's strategic blueprint?
I think it's fair to say that a year ago, if you were to ask 10 CIOs, three of them would say they had a Linux strategy in place, two or three would have heard about Linux or thought about it but not done anything about it. Today, there's not one who would have no story about what they are going to do with Linux. They have at least considered it.

The CIO of eTrade recently related a story that he had migrated his applications off Solaris onto Linux and for every $30 he used to spend, he spent $1 today. I was reading it and saying 'there is not one CIO in the world who would question whether that would apply to them.'

In the current economy, the momentum of Linux has been enhanced, not slowed, by the fact that there is such a crunch in budgets. Virtually everybody is looking at this and saying 'we have to do this.' So, what's holding back Linux adoption in the enterprise?
One of the barriers to get across is application availability. I can't tell you that we are 100% there, but we are a long way there.

Application availability is the biggest barrier. Our sales force comes back from customer engagements and tell us, 'I could succeed if I had availability of this.' Today, I don't think this is the major hurdle any more. That is because we are 60%-70% there. Any of those 30-40% are noticing today that their customers are moving to Linux and if they are not there with applications, someone else will be.

The other one is the more conservative companies want to hear from other deployments and whether they are working. I think the references are there now.

There's also questions about performance metrics. I think there are benchmarks out there now that validate that Linux is not only a solid replacement for Unix, but generally a better performer at the same price point. This is true across TPC for the database industry, Domino, application benchmarks. All of those benchmarks are in the public release. We now get into the business of optimizing it. What is the general tone of conversations you have with potential enterprise customers regarding Linux?
Nobody's debating Linux momentum in the enterprise. A year ago, I was having to defend myself saying this is a happening thing. There are the obvious drivers like the cost benefits of moving from proprietary Unix on expensive machines to an Intel platform. We routinely see customers realizing 10x price-performance improvements. That's on the low end of the scale in many cases. It's easy doing business that way. If application availability is close to becoming a non-issue, what's the next hurdle enterprises will need Linux to clear?
We are now saying that the next area of attention is going to be systems management. If Linux becomes a viable and widely deployed platform in the enterprise, you need to have support and systems management capabilities, otherwise that's going to be a barrier.

If there is a company that deploys 200 Linux servers and it is their demand to put it in one rack or one room, they need to know how to do that. One of our customers, Papa John's pizza, has 1,500 different locations. For them, it's not a matter of putting a new CD in your card. Things need to be deployed in an automated fashion. Administrators need to see if these systems are operational. The broader the deployment, the more management they need.

I think it was IDC [International Data Corp., Framingham, Mass.] that said of enterprise customers, 35% are in deployed mode and another 28% were in trial mode. It is our observation, that the people who have not deployed are our biggest opportunity. Often, the result of their initial deployment is better than they expected. We have a very easy time selling to banks, for example, that have deployed in the back office and now need other applications.

Click here for part two of's interview with Red Hat.


Best Web Links on decision criteria

Share your thoughts on this story and other enterprise Linux issues inside the discussion forum

Dig Deeper on Linux servers

Start the conversation

Send me notifications when other members comment.

Please create a username to comment.