* Metro Ethernet MAN services redefining the notion of "fast"
Historically, access links have been the great bottleneck for wide-area networking, and, by extension, these same access links have also limited the feasibility of metro networks. Since the transmission technology from the premise to the central office has been limited to copper – and occasionally fiber – the speeds for these links have been more along the traditional WAN speeds of 56K bit/sec to T-3 (45M bit/sec) while LAN connectivity has exploded from 10M bit/sec to 100M bit/sec to gigabits per second any beyond.
Now, however, metropolitan-area networks (MAN) based on metro Ethernet services are redefining the preconceived notions of what “fast” means. At the same time, the availability of what was previously considered extreme bandwidth is making a significant impact on defining the importance of various networking functions.
This fact was driven home for Steve recently when he spent some time with the Charlotte-Mecklenburg Schools in North Carolina, one of the larger school systems in the country. The system, like others across the state and across the country, is quickly moving to interconnecting the schools and supporting offices via metro Ethernet, essentially resulting in “one large LAN” for the entire system.
In the case of Charlotte-Mecklenburg, by mid-September the school system will have converted all of its connectivity from frame relay (using 256K bit/sec fractional T-1) and a scattering of NMLI (Native Mode LAN Interface) services to BellSouth’s metro Ethernet service. High schools and some administrative offices will have 100M-bit/sec connectivity with the ability to burst to 1G bit/sec. Elementary and high schools will have a 10M-bit/sec connection, with the ability to burst to 100M bit/sec.
Admittedly, this connectivity does not leave the county at these speeds. However, as noted in the most recent newsletter, this does provide the infrastructure for accessing high-speed MPLS services and other similar services for the wide area without the limit of getting to the WAN through a very narrow pipe.
And, by the way, the monthly cost for this service is actually less than the prior traditional services. The affordability is a key point. The big pipes have been available for a while, but we’re not approaching affordability for more users.
This ultimately will redefine many of our traditional ways of thinking about traffic management. Traditionally, we’ve often had to apply strict traffic management to deal with the bottleneck between massive bandwidth on the LAN and semi-massive WAN bandwidth. But the exact impact on traffic management is definitely a “bigger than a breadbox” topic, so that discussion will be deferred to a later date. Suffice it to say that in a few years, we won’t be in Kansas any more.
