At the recent Gestalt IT Wireless Field Day event, access point maker Aerohive set this up as its central differentiator: While most established wireless access points use a single high-powered central controller, Aerohive distributes the workload evenly across all access points. This same architecture was discussed by HP and has been adopted by others in the wireless access point space. By exploding the core, Aerohive promises that they can scale more efficiently than competitors. Although I am not a wireless networking expert, this discussion brought to mind many of the arguments put forth by cloud computing proponents. The core message is that centralized resources can only scale according to Moore's Law, while distributed resources can surpass it.
A similar message came through this week from data center switching vendor Force 10 Networks. Its new Z9000 Distributed Core System is similar in concept to the Aerohive "no controller" architecture, in that no central device controls the network and all participants share the processing workload. But it goes beyond this: Both systems are also optimized for
East-West traffic. Traditionally, data center networks have focused on North-South network traffic. The assumption is that clients on the edge would mainly communicate with servers at the core, rather than across the network to other clients.
But server virtualization changes all this, with servers, virtual appliances and even virtual desktops scattered across the same physical infrastructure. These environments are also highly dynamic, with workloads moving to different physical locations on the network as virtual servers are migrated (in the case of data center networks) and clients move about the building (in the case of wireless LANs).
Although some might object to the use of the word "cloud" to describe these new flexible networks, the metaphor seems to work. Rather than a hierarchical and static network, this new architecture is flexible, open and dynamic. And, as has been shown in the cloud computing space, distributing tasks achieves greater scalability than applying ever increasing compute horsepower.