- IBM employees buzzing about layoff rumors
- AT&T builds $23M IPv6 network for U.S. military
- Outlook '09
- Is VoIP dead?
- Microsoft layoff rumors continue their swirl
With 15,000 employees and revenue topping $2.7 billion, Booz Allen Hamilton is one of the largest and most successful business and IT consulting firms in the world. This fall, Booz Allen is replacing an old frame-relay network from MCI with a Multi-protocol Label Switching infrastructure from AT&T. Daniel Gasparro, chief technologist at Booz Allen, spoke recently with Network World Senior Editor Carolyn Duffy Marsan about how to select a viable carrier and why he's bucking the trend toward Internet telephony. Here are excerpts from their conversation.
What is your role with regard to the company's global network infrastructure?
My group, which is Operations Services, has about 160 people worldwide. We have nine service offerings that we deliver to the firm: infrastructure; end-user computing; security; telephony; video; e-mail; collaboration; data management; and global business applications, including human resources and finance. Our mission is to support these services worldwide.
You awarded a three-year, $5 million contract to AT&T this summer for the new MPLS network. What are you trying to accomplish with this upgrade?
We want to simplify the number of technologies that we have deployed for our data network. Before, we had frame relay. We had point-to-point connections. We had DSL. We had VPN connections. So we had a hodgepodge of technologies. We are going to standardize the majority of our connections on MPLS. We are not converging our networks - that is moving voice and data over a common IP network. We've been pretty explicit that we don't want to go to voice over IP. We are going off of frame relay altogether. The reason is that we can do quality of service with MPLS that we couldn't do with frame relay.
What carriers did you consider for this upgrade?
We looked at Equant, Sprint, AT&T and Verizon. We also looked at having a mixed portfolio. We looked at the technology from each carrier and the risk. By risk I mean, are all of these vendors going to be around three years from now? Oddly enough, once we got down to the short list, there was not a whole lot of difference in terms of service delivery. They all had points of presence where we needed them. They all had MPLS. It came down to the customer service experience and how they had done with the quality of service they have been delivering to their current client base. We started asking questions about their strategy, which is not something that a lot of these procurement processes go through. We looked at how their infrastructure was designed, their reach around the world, who they were in partnership with and what was the long-term viability of the firm. Ours was a much more comprehensive process than I've seen other firms go through.
Having done all that analysis, what is your take on the state of the telecom market right now?
I don't think it's ever going to be settled. We're going to see some carriers surge forward because of the wireless buildup, which has been overshadowed in the last five years because of the carrier problems. It's refreshing to hear that certain carriers are now offering quality of service for wireless services. All the carriers are wrestling with how they can serve business better.
What were the top business drivers for your network upgrade project with AT&T?
The No. 1 driver was availability. And not just from a technology perspective but from the perspective of company viability. What you don't want is to find out that you're working with somebody who is suddenly going to be diverted because they have financial problems. In the not-too-distant second place was the question: Are they able to grow with us in terms of adding new services without adding significant cost? In the old days, if you added two offices, you saw your bandwidth and your costs jump by an equal percentage. In terms of my total operating budget and how it grows as a percentage of revenue, it has to be very controlled growth. When you're running an overhead business you have to be smart in how you apply those dollars.
What are the bottom-line benefits that your new MPLS infrastructure will provide to your consultants in the field?
Better network availability. It isn't so much network outage related as it is performance-related availability. The networks seemed to clog up after a while. So there is better performance and redundancy for offices that are considered hub locations. We didn't really make any gains financially because we didn't have back-up capabilities before and now we do. We got more capability for about the same cost.
Will any particular applications perform better on MPLS?
I don't know how much performance gains I'm getting with MPLS because I'm converting our No. 1 application over as well. We're switching from Netscape to Microsoft [for Web and messaging applications] right now. I read, and our technical staff read, that MPLS is a more forgiving network technology when you're oversubscribed. I am hoping that is the case. Right now we're going from the old network to the new network. We're not even taking a breather, and we're adding a whole new application suite. And it is our critical application at that.
Comment