Inktomi CTO outlines wireless, enterprise plans

What are Inktomi's plans in the wireless arena for the coming year?

Inktomi has been dabbling in the wireless area for quite a while. We are monitoring the market, and we have what we think are a set of key infrastructure products. When you look at the types of services, like transformation and other kinds of optimization you want to do to Web content for handheld devices, our Traffic Server platform, and especially the ability it has to give users plug-in interfaces, provides a great platform for wireless infrastructure.

We have done trials, we've talked to wireless carriers and we continue to monitor the market. When we start to see the market pick up, we'll invest more resources in it. We won't sell applications that users are going to download and see on their wireless devices. We will sell software technology that is embedded in our partners' hardware.

Has the idea of content delivery network (CDN) peering fallen by the wayside?

I think there is a timing element to this issue. There is the Internet bubble and all the hype that surrounds it. And then there is the reality, especially given the current economics going on, of the pace at which the relationships and the business models can be worked out. From a technology perspective, Inktomi has the most advanced capabilities, especially on the streaming side, to be able to interconnect different networks together to do peering at the content layer and actually make it happen. And our customers have done this with our technology quite successfully for very large events. So we've proved the technical concept.

The challenges are more around the business models and moving forward as this space matures. As we start to see more of the studio and record labels get comfortable with bringing their content online, those business relationships will get worked out. And there is no doubt that the underlying ability to peer at the content layer is a necessity, and it will get worked out. It's just a matter of time.

Do you think CDNs will ever be as ubiquitous as the public telephone network?

In the sense that there's a phone in every home, there will be a presence in every home for video-on-demand services through music download services and so forth that reach into every home. Content peering technologies will provide the technology for those services. If you think about having relationships much like the telcos have, it's compelling. What every one is struggling with is how to get there.

How will Inktomi avoid some of the issues encountered by Akamai and Exodus?

From a financial perspective, we're in a great position. We just raised another round of financing, and that's all public on our Web site. We've got enough money in the bank, and we have a fully funded business model. Unlike Akamai or Exodus, we didn't need a huge capital outlay to invest in services, hosting centers and data centers. We are a software company and we sell software licenses. The plain vanilla cache is maturing and becoming more of a commodity. But the networking together of cache technologies at the edge of the network and how you manage them and get content to them - and how you manage bandwidth between points - is not. All the technologies that you need to make that work start to fit into applications like streaming media, corporate communications and e-learning. We have an architecture that is arguably a year ahead of our competition to make that work.

What challenges do you face in being a software-only company in what may become an appliance and blade world?

We have been successful through strong partnerships with hardware vendors. We have signed up HP, Dell, Compaq, F5 and 3Com as vendors that take our intelligent software and package it up in an appliance form factor to sell to the enterprise or service provider. That's been very effective. And as long as we continue to innovate and build value on top of their platform, we are going to be in the best position as the owners of that software.

Traffic Edge is the next generation of our content caching platform. It has media capabilities, content delivery capabilities and a pretty sophisticated management and monitoring piece. That's the thing that a lot of our OEMs see as a very strategic play, because if this architecture plays out, there's going to be a whole new market and all these enterprises are going to want to buy a new box and stick it in their branch office. Once you have that presence, you need to manage it. But in creating this rich distribution method, you solve the problem for the enterprise of getting content out to users and managing how it gets there.

Over the next year, how much will users see in the way of combined, integrated offerings that weave content and traffic management into a single box?

Connecting the user to the right edge in the Internet at large is an interesting problem that some people are trying to solve. We have focused more on problems like how the system adjusts to the various bit rates of traffic flows to give the CEO's Webcast priority delivery over something like a document distribution job. We have a platform that allows IT to set policies and control how the network behaves.

How much of an emphasis will Inktomi place on the enterprise vs. the service provider arena?

We believe in both markets. We think that today, there is a growing opportunity in the enterprise. There is this whole edge of the network in the enterprise that we want to go after. We are still investing in the service provider technology and think that in the long run, it's a viable market. The revenue opportunity in the service provider space and the short-term growth opportunity in the enterprise - given the trend toward wanting to do decentralized work and the difficulties with traveling - are both important.