The Enterprise GENI Team

The Enterprise GENI team at Stanford University is working on a project to demonstrate how GENI experiments can be a part of local networks, such as campus and enterprise networks, while simultaneously running production traffic. The Enterprise GENI team, led by Nick McKeown, Guru Parulkar, and Guido Appenzeller has successfully demonstrated commercial vendor’s switches and routers running their own implementations of Stanford’s OpenFlow protocol on the Stanford campus, in Internet2, and in JGN2plus in Tokyo, Japan.  OpenFlow supports multi-layer network slicing, creating virtual network slices that can be defined by any combination of physical layer, link layer, network layer and/or transport layer flow rules. The team is currently adding the GENI Aggregate Manager interface to the OpenFlow FlowVisor to enable GENI experimenter access to OpenFlow environments along with enterprise access to GENI.  Over the next two years the project plans to run additional campus trials with other schools and to develop a kit that will enable other researchers to replicate Enterprise GENI’s work on their own networks.

The OpenFlow Network Virtualization Software consists of two components: a FlowVisor and an Aggregate Manager. The FlowVisor virtualizes a physical OpenFlow switch into multiple logical OpenFlow switches, which can be controlled by different experimenters and is critical to allowing multiple experimenters to run independent experiments simultaneously in one physical campus network. The current HyperVisor implementation has successfully run five distinct research projects, each in their own network slice. The GENI Aggregate Manager is an OpenFlow controller that can control a subset of the network resources as defined by the local administrator.