Joel Trammell, CEO of NetQoS will be producing the welcoming address at NetQoS Symposium 2008. We asked him a few quick questions about what he’ll be talking about when Symposium starts April 20, 2008.

NPD: This year’s welcome address is called: "Why networks fail and why the role of the network engineer is secure." Could you tell me a little bit about why you chose that topic?

Trammell: Well, I think with fault management issues, people really understand the network up and down. It's very clear to them the value that keeping the network up provides. With performance, it's somewhat less clear sometimes, to people. What problems are people actually solving? What causes these performance issues? They tend to be more nebulous in nature than fault issues.

NPD: So when you say "Why networks fail," you mean "Why do networks fail to meet goals?"

Trammell: Yes, "why networks fail to perform," is really the title here.

NPD: How do networks fail to perform?

Trammell: Well often it's built around changes in the environment. There are a lot of things going on in the environment these days that add a great deal of complexity, whether it be the introduction of new applications such as VoIP, whether it be different uses of the network with folks expecting anytime anywhere access to the network, so therefore instead of just having people sitting in a building on a wired connection, you may have a wireless connection or at their homes, or at hotels, or wherever they're coming into contact with network services.

There are new technologies being deployed in the network, WAN optimization being one that has been particularly hot in the last few quarters. So all these changes introduce many opportunities to cause the network to behave in a way which is different from what the users have come to expect.

NPD: The other half of the topic is "Why the role of the network engineer is secure."

Trammell: So, yes, there's one school of thought out there in the industry that IT and particularly networking as an area will become a pure utility in the near term and that companies will no longer invest in networking, just like they no longer invest significantly in their own power distribution systems or their own power generation systems. I don't believe that's the case. I don't believe, when you think about the network - is the wild west. There are all kinds of applications being introduced, both intentionally and unintentionally on that network. Lots of technology changes going on - it's a very wild environment. It's not an environment that's conducive to a utility type approach. And therefore people with expertise in networking, and particularly performance of applications across that infrastructure will continue to find the field very lucrative.

NPD: One thing that gets me is that when people talk about the idea of the network becoming like a utility - they often bring up, "Well, people don't have their own power plants. But, they do have their own power plants - if you're a hospital, you have a power plant. If you have a data center that needs 100% uptime, you probably have at least a backup power plant.

Trammell: And people often choose to locate their facilities near where power supply is good and cheap, so even though it's a utility, geography doesn't have equal access to power at the same costs. So even power is not as pure a utility as a lot of homeowners may think of it, as when you get to use it on an industrial and commercial scale. And I think networking adds at least another layer of complexity to that.

NPD: There are significant changes of course; there is truth that Software as a Service does change a lot of things. Obviously you can't ignore that. But you still need a network to access SaaS apps.

Trammell: SaaS assumes, as sort of a priori knowledge and capability that you have this highly performing, ubiquitous network available to deliver SaaS. If anything, it makes the network even more important, because now, the only way to access the software is through the Wide Area Network.