Traffic has changed so must your Infrastructure
May 11, 2010 Leave a Comment
Network traffic profiles have changed. Back in the 1990′s email connectivity to a desktop was one of the primary driving forces for the deployment of Local Area Networks, LANs, built primarily with Ethernet. Lets look at the traffic characteristics of e-mail…. non-real time, asymmetric, bursty traffic pattern that goes from a fixed, fat client to a fixed server and back. This is the traffic pattern that defined today’s networking architectures.
In order to better support the delivery of email to more users cost effectively, the designers of network equipment built in over subscription; basically to deliver more ports, at a lower price, but each port will never have full bandwidth capability. This was perfectly fine for the e-mail application and many devices were oversubscribed at 4:1 or more and deployed in a classic Access – Distribution – Core model. This is essentially 64:1 over-subscription in each direction – again not a problem for a bursty, non-real-time and asymmetric traffic flow. We all assumed VoIP would change LANs, but a 64kb/s traffic flow doesn’t get congested too often on a GbE LAN and if it does some simple QoS prioritisation can deal with it.
But the adoption of Web2.0 applications will kill your Infrastructure….. If its Social Enterprise, SOA, VM, Grids, Clouds – public or private your data no longer goes North South, in fact it travels in every conceivable direction and as you move towards a Cloud based application model your nice local predicable data is now various IP hops away over your 64:1 contended network… And as all network managers know jitter and latency kills applications.
So, in summary, today’s applications require a new infrastructure, a 2.0 Infrastructure – low latency, low hops, flat and simple network. If you don’t have one, best you get one…..


Clearly Tarmacadam. It has a few years on Ethernet. The first Tar road was laid in Bagdad circa 8th Centaury AD, a few years before Robert Metcalf and David Boggs published their paper in 1973 outlining a successful Multipoint shared Network running locally at 3Mbp/s. 40 years later, Ethernet Networks are delivering 10Gbp/s allowing CIOs to adopt virtualisation, rich media and ignore network quality as bandwidth is now so plentiful. While LANs race towards 100Gbp/s the network connectivity that needs to go beyond the LAN has struggled to keep pace, actually local traffic has never been an issue for most, it’s the WAN that’s still expensive and slow