The specification is available for download: openflow-spec-v1.1.0.pdf

More information about the release is found in the announcement below:

Dear OpenFlow Community,

On behalf of the many people who made it possible, I am delighted to announce that “OpenFlow v1.1 (Implemented)” is now complete, published and posted at http://www.openflow.org/documents/openflow-spec-v1.1.0.pdf. The new version adds significant new features to OpenFlow, particularly for multipath, tags/tunnels, and the ability to exploit multiple hardware tables in chips.

The specification was developed over several months by a group of very dedicated people at the weekly OpenFlow meetings. I would like to thank everyone who attended and contributed. I would like to particularly thank and congratulate the working group for the huge amount of work they put into leading the discussion and writing the specification through many iterations. The working group included (in reverse alphabetical order): Jean Tourrilhes, Rajiv Ramanathan, Brandon Heller, Glen Gibb, Ed Crabbe, and Martin Casado. Jean deserves special thanks and credit for bravely stepping in to lead the team and the discussion in the later stages.

After the first release “OpenFlow v1.1 (Unimplemented)”, the community spent several months implementing each new feature, and testing it. The details are on the OpenFlow wiki for everyone to see. This involved creating a new Python OpenFlow reference switch (ofps) and associated tests (in oftest), as well as some prototyping in Open vSwitch. Many people contributed to this effort from around the world. I’d like to particularly thank: Dan Talayco, Rob Sherwood, Masahiko Takahashi, Tatsuya Yabe, Zoltán Lajos Kis, Dave Erickson, Saurav Das, Casey Barker, Bob Lantz, and Yiannis Yiakoumis, as well as additional developers at Ericsson and NEC. I apologize for anyone inadvertently missed from the list. It was a huge amount of work, as well as a huge coordination effort. Congratulations for getting this all done.

“So what’s next?”
At this point, the specification work for OpenFlow v1.1 is complete, although we do not want to rule out the possibility of a “OpenFlow v1.1 (Integrated)” specification, after a fully integrated implementation is complete. But at this point, we believe the current specification is sufficiently stable for vendors and users to start development work.

As many of you know, we will be moving the OpenFlow specification effort into a newly formed standards group soon. This will be announced at the end of March (or early April), and we will be sure to announce it here. The new group will drive the specification of future versions of OpenFlow, beginning with OpenFlow v1.2. If you have general questions, hold off for a couple of weeks and things should get clearer. If you would like to volunteer to be closely involved, please email me.

Instead of weekly OpenFlow meetings, we will hold discussion meetings about every 2 weeks, with the first one on Wed March 16 at 4pm. (More details to follow). We would like to get input from folks involved in the process, to help us set the stage for future versions, including gathering together some features we need to include (e.g. IPv6, more general matching,…). If there are topics you’d like to discuss, please email Glen Gibb and he will keep track of an agenda and calling meetings.

Watch this space. This is still only just the beginning!

Onward and upward,
- Nick

The next OpenFlow tutorial will take place on August 20 in Mountain View, CA as part of HOT Interconnects Symposium.

This tutorial is an opportunity to gain hands-on experience with the platforms and debugging tools most useful for developing network control applications on OpenFlow. Following an introduction, each participant will create a flow-based Ethernet switch. Along the way, attendees will learn the full suite of OpenFlow debugging tools: view flow tables with dpctl, dissect packets with Wireshark, slice with FlowVisor, simulate a multi-switch, multi-host network with Mininet on their laptops, and control a network with real switches.
The only requirement is to bring a laptop; no experience is required.

For further details regarding HOT-I 2010 and registration, click here !

Unlike previous releases, OpenFlow 1.0 is the first release of the standard that we feel is ready to be implemented in generally available products. We have been running OpenFlow in our Stanford network over the past months, and it has proven itself for both production use as well as experimentation.

OpenFlow 1.0 adds a number of key features. The largest addition is Slicing, a simple QoS mechanism that allows the isolation of traffic in OpenFlow networks. Smaller changes include matching IP addresses in ARP packets, Flow Cookies to identify flows, selective port statistics and matching on the ToS bits in the IP header. The release also includes a large number of small changes to the specification and bug fixes in the reference implementation. For a complete list of changes read the Release Notes or the more detailed Wiki Page.

Hardware accelerated OpenFlow 1.0 capable switches are expected to become available over the next months and we will keep you updated on the OpenFlow Blog and the OpenFlow-Announce Mailing List. On the controller side the reference controller supports 1.0 and there is a version of NOX with partial support available. If you deploy OpenFlow 1.0, feedback and bug reports are highly welcome via Trac or the openflow-discuss mailing list.

This release would not have been possible without the work of (in alphabetical order) Guido Appenzeller, Peter Balland, Martin Casado, David Erickson, Glen Gibb, Brandon Heller, Mikio Hara, Bob Lantz, Masayoshi Kobayashi, Nick McKeown, Justin Pettit, Ben Pfaff, Rob Sherwood, Srini Seetharaman, Dan Talayco, Jean Tourrilhes, Tatsuya Yabe, KK Yap, Yiannis Yiakoumis as well as to the many other members of the OpenFlow community that provided valuable suggestions, feedback and testing.

Thanks to all of you!

Brandon, Glen and Guido

The features described in the tech report have been included in the newly released FlowVisor version 0.4, including:

* preliminary bandwidth isolation between slices

* improved OpenFlow message rate limting for switch CPU isolation

* per-slice virtual topologies

Both the technical report and the new version are available from the FlowVisor’s webpage:

http://www.openflow.org/wk/index.php/FlowVisor

OpenFlow 0.9 is a major step towards the 1.0 release, and it incorporates a large number of bug fixes, clarifications and new features including  failover, emergency flow entries, VLAN Priority Matching and support for ToS field re-writing. For a complete list of changes read the Release Notes or the more detailed Wiki Page.

For OpenFlow deployments, we recommend to use OpenFlow 0.8.9 for now. Nox, the NetFPGA and other OpenFlow Hardware at this point is not yet compatible with 0.9. We will post on the OpenFlow Blog as these get updated. Feedback and bug reports are highly welcome via Trac or the openflow-discuss mailing list.

This release would not have been possible without the work of (in alphabetical order) David Erickson, Glen Gibb, Brandon Heller, Mikio Hara, Bob Lantz, Justin Pettit, Ben Pfaff, Jean Tourrilhes and David Underhill, Tatsuya Yabe, Yiannis Yiakoumis as well as to the many other members of the OpenFlow community that provided valuable suggestions, feedback and testing.

Thanks to all of you!

Guido, Brandon & Mikio

The new lists are:

• OpenFlow-announce — a low-volume list for OpenFlow announcements including new versions of the specification and reference design releases
• OpenFlow-discuss — a list for general discussion of OpenFlow related topics including deployment, use, applications, controller development