Use Internet2 SiteID

Already have an Internet2 SiteID?
Sign in here.

Internet2 SiteID


Megaconference v6 2012 Results

Jun 19, 2012, by Ben Fineman
Tags: Audio Video Communications Infrastructure, bad_text, Collaboration, imported, Internet2 Video Exchange, IPv6, Network Infrastructure, Video Services, Video, Voice & Collaboration

The day began early for us in Ann Arbor. I arrived shortly after 6am Eastern time, laden with doughnuts and coffee to fuel my colleague and local video technician for the day, Jeff Hagley.

After flipping a few switches, the gear the we had deployed the day before hummed to life. I settled myself in from of the green screen, selected the long IPv6 address of the MCU from my recent calls, and one ring later we were online.

Megaconference v6 at Internet2

Despite the early hour, the event itself was a pleasure. I'm always impressed and humbled by the outpouring of community support to promote collaboration in the research and education community. This event was no exception. We started the call with greetings from our follow video engineers around the country and the world.

We had participating U.S. bridges from OARnet in Ohio, MAGPI in Pennsylvania, CEN in Connecticut, and the State of Minnesota, in addition to the Internet2 Video Services bridges. On the global front, there were bridges from AARnet in Australia, ARNES in Slovenia, and the University of Helsinki in Finland. The bridges were all either Cisco/Tandberg/Codian or Polycom, and while all the bridges were "IPv6 capable" only eight of the ten had IPv6 working due to local technical issues.

At 8am Eastern time, the event officially kicked off and participants began to join. I'd like to extend kudos to those that took the time to dial in, especially those who made it using IPv6. We had around 50 participating sites from around the world, with about half successfully connecting with IPv6. Endpoints were primarily a mix of Cisco/Tandberg, Polycom, and Lifesize, but also including a smattering of softclients and a Grandstream phone.

Those that joined us were treated to a lively and exciting program, including author Vicki Cobb, students from the National Dali High School of Taiwan, Ann Hernandez from the Ann Arbor Hands on Museum, Cindi Collins from the National Cowgirl Museum, Michelle Cruz from the Mount Washington Observatory, Megaconference founders Bob Dixon and Sandy Sprafka, Rosa Catala with live surgery from the Library Science Center, Galen Sjostrom with live animals from the Minnesota Zoo, Lee Gambol from the Cleveland Museum of Natural History, Jeff Hagley and Ryan Martin from Internet2 on Eduroam, and Shumon Huque from the University of Pennsylvania on their IPv6 deployment. I could not have hoped for a better set of presenters - their content kept me not only awake but also interested as the day wore on.

As it got later in the afternoon and Europe was following Asia into the bedtime hours, things quieted down a bit, but we enjoyed some good conversation. I was pleased to have some of our Australian colleagues rejoin us in our evening as their morning began. We ended the day on a high note at 8pm Eastern time, exhausted but fulfilled.

Some overarching observations: IPv6 still has a long way to go in the video world, but we continue to make progress. Almost all modern endpoints and bridges now come with IPv6 support in the most basic sense - IPv6 can be enabled, and calls to IPv6 addresses can be made, in most cases. In some cases, there are bugs that continue to emerge in uncommon infrastructure environments that were not previously uncovered simply due to the relatively small amount of units using IPv6 today. Some endpoints, often soft clients, simply do not work with IPv6 yet. This has historically not been a big issue as home network did not have IPv6 anyway, but this is becoming less true - we had more than one participant coming in from their home using IPv6.

A very important area that continues to be a challenge is video infrastructure. Some infrastructure is still lacking IPv6, which causes the endpoints they support to lack IPv6 (such as some soft clients, CTS, others). While many gatekeepers and SIP registrars do have IPv6 capabilities, manufacturers need to improve the way endpoints interact with them - in some cases endpoints cannot register using IPv6, or when they are registered via IPv4 they cannot call IPv6 addresses. On the other hand, infrastructure is progressing in such a way as to make IPv6 interoperability more seamless - IPv4 to IPv6 traversal can mitigate the lack of one protocol or the other, and the proliferation of URI and E.164 dialing holds the promise of removing the need to dial those long IPv6 addresses (or any IP addresses, for that matter). Participation in Internet2's interoperable dialing infrastructure, or that of our global peers, helps organizations progress beyond IP dialing.

In conclusion, I am heartened to see the progress that we continue to make. The research and education community has a strong history of coming together to make the technology work (and tell the manufacturers what we need for the technology to work) to do the things that we do best - collaborate, and help each other reach our mutual goals.

View the video archive of the event at