When the infant OpenFlow protocol began attracting notice outside of research institutions last year, there was a great deal of excitement, but often a bit less understanding of what it actually supports and entails. Here is how the Open Networking Foundation (ONF), the group responsible for the development of the OpenFlow spec, defines it:
OpenFlow is an open standard that enables researchers to run experimental protocols in the campus networks we use every day. OpenFlow is added as a feature to commercial Ethernet switches, routers and wireless access points – and provides a standardized hook to allow researchers to run experiments, without requiring vendors to expose the internal workings of their network devices.
Brocade understood the possibilities of Software-Defined Networking (SDN), including OpenFlow, early on, and became a charter member of the ONF, helping to define the OpenFlow spec and working to support its commercial implementation.
Last week, Brocade participated in the ONF’s first OpenFlow Plugfest, with a CER and an MLXe supporting a pre-alpha version of the OpenFlow feature to be released as a GA product later this year. By the way, it’s important to understand that the main purpose of the OpenFlow plugfest was to test interoperability between OpenFlow controllers and OpenFlow-enabled switches, since there is no protocol interaction between OpenFlow-enabled switches. (For more detail on this, see the comments section of Michael Schipp’s blog, Is OpenFlow Losing Its Openness?)
At last week’s event, Brocade demonstrated support for OpenFlow with traffic forwarding in hardware for wire-speed performancewith 1G and 10G interfaces. The MLXe also supports OpenFlow with wire-speed performance for 100G interfaces. However, the plugfest was limited to 1G and 10G interfaces only due to other vendors’ limitations. All of the controllers were able to successfully connect to the CER and MLXe. The controllers tested included NEC PFC, Indiana University ON-SS, NTTData, BigSwitch, and FlowVisor. It was the first time the CER and MLXe were connected to the NTTData, BigSwitch, and FlowVisor controllers. All the controllers were able to discover the topology of the network connected to the CER and MLXe, and all traffic tests involving the CER and MLXe were successful.
We look forward to sharing more of our thoughts and work around OpenFlow and SDN in the coming months!