Building the perfect education and research network

After a lengthy process of gathering requirements, procurement and design, during 2013 Janet completed the build of and migration to its new backbone network. It is, however, not perfect. This talk will look at what might constitute a perfect network for research and education, and why this holy grail is not achievable, and perhaps not even desirable. The talk will cover some of the design choices that must be made when designing and building a network, either in the name of cost, or of feasibility, or of a lack of demand from users (imminently or forseeable), or simply in the name of creating a reliable service. It will start by describing a possible perfect NREN architecture, then move on to describing, at a high level, the architecture of Janet6, highlighting the differences, and why those choices were made. After that, the talk will cover some of the ideals in more detail and analyse other reasons why those features, as desirable as they may be, may or may not make sense to deploy.


Part of session

Networking architecture

Related documents