ility to accommodate evolutionary rather than revolutionary change.
Plot a course and stick with it.
Don't lurch from technology to technology just because you happen to read about something
new. "Some managers decide they need frame relay immediately, even if they don't really know what it is" or how it might benefit their company.
Don't expect a quick fix from any technology.
"Customers often apply new technologies to their existing business processes and obtain mixed results. You must change your business processes to take advantage of new technologies."
Invest in ATM to the desktop.
If you've prepared your business for fundamental change, don't hold back making the move to ATM. ATM will help you cope with customer expectations for network performance. "People who want Microsoft Exchange to operate within 10-Mbps Ethernet today will want 100BaseT the year after that. The year after that, 100BaseT won't be adequate. In contrast, ATM to the desktop should satisfy customer demands for the next five years. Furthermore, if you put in ATM to the desktop, you won't have to change the wire-closet switching fabric."
Take matters into your own hands
when necessary.
Many tariffed carrier data services are either too high-priced or inconvenient from a customer perspective, which leads companies like EDS to build a private ATM network to service its customers. "Carriers like AT&T and MCI will accept only 53-byte cells [ATM data packets] from their customers, no exceptions. Therefore, customers are required to do all data conversions prior to accessing the network. On a scaled basis, it might cost me $4.60 to adapt my SNA and frame relay traffic to the DS-3 format required by a carrier to connect to their tariffed ATM backbone. [It would cost] $1 if I had stayed with plain old DS-3."
Convergence Do's and Don'ts
