Over 280 leading networking researchers from around the country gathered this week for the 10th GENI Engineering Conference (GEC).  The meeting — co-hosted by the Polytechnic University of Puerto Rico and the University of Puerto Rico — came at a critical time in the evolution of the GENI Project:  out of the “startup phase,” the GENI Project Office (GPO) is seeking to substantially ramp up experimentation in the coming year, all the while enhancing build-outs in campuses and backbones throughout the nation as part of the growing meso-scale GENI.

(As we’ve blogged in this space before, the GENI Project was first funded by NSF in 2007 — to take a clean-slate approach and create a virtual laboratory for exploring future internets “at scale,” all the while opening up new and exciting areas of research at the frontiers of networking science & engineering.)

The structure of this week’s conference illustrated the tremendous progress the GPO and countless research teams working side by side have made in just the past year, let alone in the three years since the NSF first launched the GENI Project.  No longer were the Working Groups that have been the focal point of the past several GECs necessary.  Instead, there were three new concurrent tracks, focused on the experimenter (helping researchers learn how to use GENI for experimentation), campus/infrastructure operator (exploring issues of deployment and operations of GENI infrastructure), and software developer/system integrator (exploring key aspects of building/integrating software to support GENI experimentation within the next 12-18 months).

Moreover, a key goal of this particular GEC was to make a series of important engineering decisions in order to prepare for “at scale” GENI.  As Chip Elliott — who has worked tirelessly as head of the GPO over the past several years in an effort to bring the GENI vision to reality — acknowledged during his opening remarks Wednesday morning, these decisions are merely “rev. 1.0 decisions.  None of us believes these will be the end-all, be-all.  We don’t have to be perfect at the beginning, but we have to make [these decisions]” in order to ensure timely forward progress.  Among the topics discussed at the meeting:  software identity and attributes, authorization, resource descriptions (rspecs), and inter-aggregate stitching, as well as basic operations agreements on GENI participation.

In addition to the tracks and decision-making described above, numerous teams demonstrated their progress as part of a Tuesday evening demo session — a growing tradition at the GECs.  There were 45 incredible demos this time around — up from 37 at the last GEC in Washington, DC, in November and 18 a year ago in Durham, NC.

Wednesday’s plenary featured two Future Internet Architecture (FIA) projects — namely Peter Steenkiste’s (CMU) effort to build an eXpressive Internet Architecture (XIA), focused on trustworthiness and evolvability, as well as Ivan Seskar’s (Rutgers University) MobilityFirst FIA, assuming mobility as the key driver for the future Internet.

Finally, the GEC kicked off on Sunday and Monday with the first-ever GENI Doctoral Consortium, which brought together an exceptional group of graduate students from the U.S. as well as Germany.  The students worked with one another over the two days, sharing ideas, brainstorming, etc.  A goal was to provide the students with an opportunity to find potential collaborators — and many walked away with new colleagues they can call upon in the future.  Organizers will attempt more of these in the future, perhaps one every year.

Be sure to check out the full agenda, including slideshows, or learn more about the GENI Project here.

And if you were at the 10th GEC, please share your thoughts below!

(Contributed by Erwin Gianchandani, CCC Director)