Convergence and Crystallization in Networking for Research and Education
Presented at the Fall Internet2 Member Meeting
Austin, Texas
September 28, 2004
Abstract
We in the world of top-level networking for research and education are in a brief period of exceptional opportunity. Forces and resources, institutions and associations, ideas and desires are converging to enable the synthesis of a new, much more potent, networking environment. If we can seize the opportunity to greatest advantage, we can realize and solidify a set of platforms that can support strong technical advances over the next decade or more. Political and institutional leaders have bought the vision. The financial resources are practically assured. All of the principal players are at the table. The question is whether optimal organizational structures will evolve.
******
I welcome you to Austin, Texas. We are very glad to have this important gathering of worldwide leaders in high-performance networking here on our home turf, and we hope that you have a wonderful time while you are here.
As president of The University of Texas at Austin, I am, in some sense, your host. We are pleased to have you in our city and on our campus, and we are honored to provide technical support for this meeting. Your colleague Dan Updegrove, vice president and CIO at UT, and his organization, Information Technology Services, have worked hard to see that you have the facilities and services that you need for success here. There are opportunities for you to visit the University, and we open our doors to you all.
Those of you who participated in the Research Channel meeting on Sunday have already experienced our Applied Computational Engineering and Sciences (or ACES) Building, with its state-of-the-art visualization laboratory and media-supported meeting rooms. All of you are cordially invited to this evening's events on the UT campus: First, a dinner at the Alumni Center, then a special performance by the University's resident Miró Quartet, capped off by a reception in the Performing Arts Center.
What makes the Miró performance special -- aside from the opportunity to hear a superb string quartet with a worldwide reputation -- is that all of the audience will experience the performance in two settings. Before intermission, half of you will be in the Bates Recital Hall with the Quartet, and the other half will be in McCullough Opera Theatre, listening to multi-channel immersive audio and viewing live high-resolution images of the performers. At intermission, you will switch halls, so that each of you can compare the real concert experience with the immersive one. Is it live, or is it Internet2? We will ask you to complete a brief survey, to increase our understanding of the prospects for delivering network-mediated concerts or master-level teaching in the performing arts over much greater distances.
******
My remarks here will be suspended in a short while by two special visitors, but let me get started with my keynote message, which is about the convergence that is happening in our area of technology.
Texas is a good site for any discussion of convergence, because it has long been a crossroads of cultures, languages, and lifestyles.
It goes back at least to 1528, when Alvar Nuñez Cabeza de Vaca shipwrecked on Galveston Island and began his seven-year journey among indigenous peoples. He gave us what might be the first written record of cultural convergence in this region. After being discovered and captured on the island by local Indians, he wrote:
"So we got to their dwellings, where we saw they had built a hut for us with many fires in it. About one hour after our arrival they began to dance and to make a great celebration (which lasted the whole night), although there was neither pleasure, feast, nor sleep in it for us, since we expected to be sacrificed. In the morning they again gave us fish and roots, and treated us so well that we became reassured, losing somewhat our apprehension of being butchered."
For Cabeza de Vaca's team, things went downhill from there. The convergence was not so harmonic.
Let's move past his story a few centuries, skip ahead past French colonial outposts, Spanish missions, war with Mexico, the struggling Republic of Texas, Statehood, the Civil War and the Confederacy, and the rest of the 19th century.
Texans are industrious people; and in the 20th century, ambition and know-how converged here. This story began with the advent of the big-scale petroleum industry and progressed toward higher technology -- through the creation of the integrated circuit (by Nobelist Jack Kilby, shown on the screen) and the heart stint, through the establishment of the NASA Johnson Spacecraft Center, and into the emergence of a world-renowned center for cancer research and treatment. Houston is still the energy capital of the world, but the largest employment sector in Texas is now microelectronics and software.
And that brings us to the host site of this conference -- Austin. In Richard Florida's recent book, The Rise of the Creative Class, he names Austin as the second most creative city in the country, right after San Francisco. And why? Because we keep Austin weird. We have become an attractive, successful city because of our lively marriage between commerce and culture:
• A center for microelectronics and the computer industry -- and a vibrant, unique music industry known around the world.
• The developers of semiconductors, software, and web-based games -- and the annual Spam-arama contest.
• Internet2 -- and Eeyore's Birthday party.
• Michael Dell -- and Willie Nelson.
That pretty much explains why Austin has become legendary. It's not afraid to be weird -- or wired -- or wireless. Or whatever . . .
******
Just as convergence produced something new in Texas -– a place with its own opportunity and creativity –- the same seems to be happening in the world of top-level networking for research and education. A remarkable opportunity lies before us. Forces and resources, institutions and associations, ideas and desires are converging to enable the synthesis of a new, much more potent, networking environment. But the opportunity is not certain to be realized. That is up to our vision, pragmatism, courage, and skill.
I speak this morning from four different perspectives.
Most immediately, I am Chair of the Internet2 Board of Trustees. In that capacity, I have a close view of current projects and operations sponsored by our organization, and I have a hand in the process that is shaping the next generation of networking.
Second, I am president of a research university -– and a public one, at that. It is my main job to see that The University of Texas at Austin remains well connected to leading institutions worldwide and that it supports the best educational and research opportunities for its faculty, staff, and students. In our capacity as a flagship in the State of Texas, we also have the responsibility to foster a superior research and teaching environment for higher education in the state at large.
I am also a scientist -– a former working chemist with a career-long interest in computing, computer-based instrumentation, and networking. Consequently, I observe the current scene in some measure from the viewpoint of the practitioner who desires sheer utility from the things we are undertaking here.
Finally, I am a citizen attentive to the health of my community, region, nation, and world, and like many regional leaders I am keenly interested in the support and opportunity that advanced networking can provide toward a better quality of life among folks at large. In this capacity, I am focused on how political leaders and taxpayers can find value enough in all of this to subscribe.
******
Here is what I see:
At the national level, there is a convergence of separate visions for new forms of networking, the agendas of key national players, and sizable financial resources.
The visions are breathtaking and encompass at least three networks of national scope that can support different needs and must have very different technical properties:
• the next-generation broadband network for intensive, general, research and education traffic.
• a network that can support active research and experimentation on advanced networking.
• a network to support very large scale national and global scientific efforts in which the network becomes part of the apparatus, like the Extensible TeraGrid Facility -- a nationwide, coordinated multi-site computing system -- or the radio astronomers' Very Long Baseline Interferometry consortia, which establish planet-spanning radio telescopes.
Remarkably, we have seen these visions converge over the past couple of years to the point that all appear to be realizable on the basis of a common physical infrastructure, the core of which would be an optical fiber network owned by the research and education community and maintained by regional and national organizations spanning nearly all major players in research and education nationwide -- major universities, private corporations, and national laboratories. The importance of this particular convergence cannot be overstated: It makes the difference between a dream and a practicality. Separate approaches to all needs would have a prohibitive combined cost, both in financial terms and with respect to the attention that enabling organizations and individuals could devote to them. By taking roads apart, we could never realize the broad reach of the current national vision, and we might not realize all three functions to any appreciable extent.
My main message today is that we have a sterling opportunity before us as a community. We must not fail to follow through. Carpe diem, indeed.
To do so, it will take the focused attention of many players. Thus, we are fortunate to have also witnessed another form of convergence: the willingness of principal organizations to move the combined vision to a prominent position on their separate agendas. I congratulate the leadership of CENIC, the Pacific Northwest GigaPop, National LambdaRail, SURA, IEEAF, Starlight, the Extensible TeraGrid Initiative, and many other groups and individual institutions for their visionary demonstration of collaboration in all of this. As chair of the Internet2 Board, I am proud of our solid, early commitment to the effort at the moment of nucleation. We dedicated $10 million to it; we have secured 5,000 miles of fiber to support it; and key individuals from Internet2 have participated in organizational activity and technical activity developing it.
******
This has been an age of convergence at the state and regional level, too. Institutions everywhere are recognizing common interests in achieving the best possible national connectivity, in anticipation of the more powerful, but more demanding, national vision. They want to be ready; they want to take full advantage; and they see a chance to achieve the results they need at lower cost through collaboration.
In state after state, the chosen path is to translate the national vision to the regional level: A membership organization is being formed to establish and to operate an owned, multi-lambda fiber optic network, covering the state and embedding the regional connection points to the emerging national fiber optic backbone. Good examples include FLR in Florida, LONI in Louisiana, and LEARN here in Texas.
In some regions, added impetus has been given to the regional convergence by connecting the vision of the regional network to the promise of addressing public interests beyond higher education and research, such as delivery of services for education, networking needs of governmental agencies, and broad economic development. The larger picture is attractive to political leaders. The governors and legislatures of several states have fostered regional networks through financial support of the kind that Governor Perry and Lieutenant Governor Dewhurst announced here a few moments ago.
It is an exciting time. One senses that something special is happening, not just nationally, but where we all live.
******
Here in Texas, we are particularly gratified by the outstanding institutional collaboration that has brought into being our Lonestar Education and Research Network (or LEARN, for short). Essentially the entire population of research-active universities (26 of them), all nine academic medical centers, and the entire Texas Association of Community Colleges have come together to create LEARN as a newly incorporated not-for-profit organization. It will hold the physical assets needed to support a multi-lambda network offering optimal global connectivity, and it will operate that network statewide. It will support the needs of both the National LambdaRail project and the Extensible TeraGrid Facility (ETF). The University of Texas at Austin will join the ETF tomorrow as one of three new partners, and several other institutions in Texas will work with UT Austin on applications of the ETF.
LEARN will also help us to remedy the fragmentation that has historically characterized networking for education and research in Texas. It has already become the principal collaborative forum for developing effective policies concerning networking within Texas; and it will become an important voice for coherent representation of Texas institutions nationally. The creation of LEARN and the enthusiastic participation of the membership is a big step forward in this region -- a convergence of a local kind, but with national implications. We are pleased to able to highlight our progress here for you today.
******
Let me now step up two geographic orders, from the regional to the international level, where there is also a convergence of interest and effort, particularly toward settling and implementing standards for operation and international connectivity. A worldwide interest in cyberinfrastructure has emerged, and the establishment of competitive assets has become a priority for developed and developing nations. In many countries, this interest extends to high-performance networking for research and education. Efforts now in motion to ensure ongoing compatibility and improved performance across national borders and under the oceans are timely indeed. Internet2's leadership, several of our member institutions, and many individuals active in Internet2 have taken a strong interest in these issues and are committed to progress. On behalf of the Board, I thank all contributors for their vision and their effort.
******
Convergence brings opportunity. Convergence produces change. We are in a very brief era in which the outline for the next generation of networking is being written. We will live for quite a while afterward with whatever evolves in the next year.
A natural question for Internet2 is the role that it can most beneficially play, either in the writing of the outline or in the time of execution and operation after the outline is settled. Doug Van Houweling will speak to that question in greater detail in a few moments, but let me express some of my hopes.
Above all, we can bring reliability and operational skill to the table. These are great strengths of Internet2, well established in the past eight years. We created Abilene rapidly and efficiently, with essentially no bad surprises; we have taken it through a couple of upgrades; and we have operated it with outstanding reliability. The sole major exception to the latter arose from the flooding of the Houston node when Tropical Storm Allison dropped 30 inches of rain there in one day of 2001. Our organization has engaged a wonderful group of practitioners, many from our member universities. They know how to design, build, and run these railroads to deliver service to members in the planned and expected manner. Their knowledge and skill will be needed -- indeed will be invaluable -- in the next chapter of development.
Second, we offer substantial convening power, as this very Member Meeting demonstrates. Internet2 can bring people together to discuss progress and to build agendas in a host of critical supporting areas, from network concepts, to hardware, to middleware, to applications, to policy. These regular, well-attended forums can be most valuable "town hall gatherings," facilitating broad-based participation in the huge effort before the networking community.
Third is the fact that our membership is already assembled and is largely the same group of institutions and individuals that will be the proprietors and clients of the new network environment for research and education.
In other words, Internet2 is a reliable, going concern. We can be counted upon to support the big and small things needed to make the next chapter work, and we are committed to serving the larger community as a steadfast partner.
******
The title that I provided for this little talk was "Convergence and Crystallization in Networking for Research and Education." I have spoken at length about convergence, but I have yet to mention crystallization.
For a chemist like me, crystallization is a beautiful, fulfilling process, the end of a synthesis, the point of payoff. It gives a result that can be held in the hand, a product that can be both used and enjoyed.
Crystallization requires convergence, but convergence alone is not enough. It does not always happen, even though the conditions seem right. The pieces, having come together, must stay together. Nuclei must form, and growth must occur because there is overall advantage in the result.
This is where we are in the world of networking for research and education. We have convergence, and we have signs that nuclei exist. Will the crystals develop? There is a good chance, if together we concentrate on sustaining the required conditions, which, in this case, include good will across our community and a clear vision of the overall advantage. I wish us all good luck, and I hope that a year from now we will have in hand the glorious sparkle of well-formed products.
Thank you so much for being a part of the effort. And thank you for listening.
Click Here for a printer friendly version.
Click Here to visit the Internet2 website.
