HPCwire: National LambdaRail President Explains Research Focus

The Leading Source for Global News and Information Covering the Ecosystem of High Productivity Computing

Leading HPC
Solution Providers

HPCwire >> Features

October 13, 2006

National LambdaRail President Explains Research Focus

, National LambdaRail

I have been asked: Why is National LamdbaRail (NLR) focusing so much on facilitating network research and "big" science applications as its core mission? Is it not as important to give equal or greater attention to the networking needs of the broader research and education (R&E) community?

In answering these questions, permit me to draw a comparison to the historical development and evolution of the great cities around the world.

When investing in real estate today, you are told that the principle focus should be on Location! Location! Location! The root of this principle is evident in the development and evolution of the major cities and regions around the globe. In most instances, the selection of the location for each of these cities was based on such strategic factors as: safety and security; access to basic necessities and resources; transportation; and, the potential to develop commerce. As each city's core developed and prospered, the community grew and expanded geographically. Over time, each great city not only evolves, but also periodically reinvents its core in order to stay vibrant and respond to the changing needs of the society. The cycle, never ending, continues to repeat itself as new needs drive evolutions.

From my perspective, the development and evolution of networking in our R&E community has followed a similar cyclic path over the past four decades with the start of each cycle focused on Research! Research! Research!

In the late 60s, the revolutionary developments in packet networks and related technologies came from university researchers. Combined with DARPA's funding and deployment of the ARPAnet, these innovations created the initial core technologies and seminal implementation for the Internet. From that era forward, both research inventions and the specific needs of the researchers have been the core drivers for the major advances in networking for the general R&E community as well as for society at large.

Inspired by the pioneering researcher-driven development of APRAnet, the beginning network needs of the broader university and lab community were addressed by a major expansion and evolution of the ARPAnet and by the implementations of BITNET and CSnet. Together these efforts were able to bring much larger numbers of institutions online enabling broad use by researchers, educators, students and even administrators. By leveraging additional platforms (in BITNET's case via what were often institutional IBM mainframes reconfigured by computer center folks to enable a few forms of network capabilities) and funding sources, these much larger numbers established fertile ground for creating and deriving the benefits of what Metcalf came to call "network effects."

In the late-1980s, the cycle renewed itself with the research community-driven convergence on the increasingly evolved core Internet protocols of the ARPAnet. Our research community created related usable network capabilities and applications like broadly usable email, ftp and listserv. And, to meet the growing needs of researchers, the use of networks in R&E became a key driver in the creation and rapid expansion, evolution and extension of the Internet protocol-based networks such as NSFnet, ESnet and DREN. As a companion to and major enabler of the NSFnet, regional research and education networks (the NSF Regionals) were developed as the most powerful means for extending the reach of, democratizing access to, and sustaining R&E networks by explicitly recognizing and addressing the diverse geographical and demographic realties of networks.

Advanced networking for the researchers and our larger R&E community suffered a serious setback in the early 1990's when NSFnet became a victim of its own success and was 'privatized' and the program discontinued. This came during a cycle in which many of the regional R&E networks and research community-based Internet providers were spun off, and either became or were sold off to the emerging commercial ISP's. While this helped enable the worldwide Internet revolution, it unfortunately left most of the R&E community without either a voice in (let alone control over) network capabilities and services, or direct access to innovating with (or configuring around) underlying network technology. The good news was that "our" Internet had become pervasive. The bad news was that we had lost nearly all control and were reduced to buying one-size-fits-all capabilities that were being evolved for mass market rather than research, education and clinical needs. And our network researchers had lost the ability to do network and network-based research and to innovate in those areas in big ways.

With the discontinuation of the NSFnet, the NSF sought to address the high-end advanced networking infrastructure needs and opportunities of the researchers with support for the vBNS program. This initiative, combined with the CoREN and other activities of the surviving Regionals, sustained the continuing regionals and stimulated their and other regional group's evolution into GigaPops. The GigaPops, in turn, worked for the creation of UCAID (the corporate entity underlying Internet2) and the development of the Abilene Network in a partnership with Qwest and Nortel. With this new cycle, our community enabled the formation of a state-of-the-commercial-art, packet-over-SONET shared layer 3 service and the ability to control bandwidth and topology. During this period, state/regional networks and GigaPops grew and developed rapidly to meet the burgeoning educational needs throughout the community.

Shortly after the beginning of this century, three new drivers emerged that called for a major leap forward in advanced networking to meet the needs of the researchers in the community and the community at large. These included:

Need One: More and more manageable bandwidth for "big" or specialized applications' research

High-end science needs terabits and petabytes
Predictable quality interconnect
Immersive (i.e. often high-resolution and low latency) presence
Deterministic control loops
Bit rate of reality-10 gigabits soon to be 40 gigabits, but more importantly given the realities of the instruments and cluster computers and storage systems many parallel 10 gigabits

Need Two: Breakable and researcher-controlled networks including waves for network research

New protocols
New devices
New architectures

Need Three: Underlying owned fiber

Ensuring that the limitations on what our communities can pursue are based on what we can imagine and afford, rather than what services telecommunications companies are willing to sell us and the conditions under which they are willing to sell them
Enabling partnering in truly new optical domains
Keeping prices from providers low

The result, National LambdaRail (NLR), has been the implementation of a fully operational national networking physical infrastructure based upon owned and lit fiber. This fiber is linked with multiple Regional Optical Network (RON) physical infrastructures that are RON-owned. By enabling the implementation of multiple networks, both experimental and production, to facilitate technology innovation, discovery and sharing of new knowledge, this new networking collaboratory is providing for the concurrent advancement of:

Network research;
Next generation network-based research applications in science, engineering, medicine and other disciplines; and,
Education program delivery at all levels -- K through 20.

By focusing on facilitating Research! Research! Research!, NLR, in partnership with the RONs, continues the network innovation cycle and ensures that all the participants in the research and education community reap the benefits of big, fast, customizable networks. The reality is that researchers are the innovators -- the folks who think big and drive myriad ways to create the networks of the future today. We in the R&E community can effectively learn from them and extend and recreate these innovative networks for universities, labs, the greater educational community as well as our society at large.

The cycle continues. What are the next drivers? Let's all learn together by serving and studying Research! Research! Research!

-----

Tom West is President and Chief Executive Officer of National LambdaRail, a major initiative of U.S. research universities and private sector technology companies to provide a national scale infrastructure for research and experimentation in networking technologies and applications. Over the years, West has served as an advisor and consultant to a number of higher education institutions and systems, private corporations and state governments. West has been actively involved in national research and education networking in the United States for nearly two decades.

Article Tools

Share & Save Options

Discussion

There are 0 discussion items posted.

Join the Discussion

You must be a registered user in order to comment on HPCwire stories.

Members enjoy:

	Access to Special Content, including Podcasts, Webcasts, and Whitepapers and more!
	Ability to take an active role in the discussions that are shaping the HPC technology discipline!
	Special offers available only to HPCwire members!

Become a Registered User Today!

Registered Users Log in Here to Comment

Email Address:	Password (case sensitive)

Remember me	Forgot Password?

RSS Feeds

Feeds by Topic

Feeds by Industry

Feeds by Content Type

Top Headlines

San Diego Gets Set for Storage Explosion

Jul 03 | Byte and Switch | The San Diego Supercomputer Center, which provides much of the core storage for the TeraGrid, is overhauling its 28 petabyte storage system to support tremendous data growth. Read more...

Intel's Gelsinger Predicts Intel Inside Everything

Jul 03 | ExtremeTech | Intel exec Pat Gelsinger said he sees the Intel Architecture permeating virtually every segment of computing, as the company's microprocessors expand into more and more cores. Read more...

A Massively Parallel Future

Jul 03 | Bangkok Post | The latest programmable GPUs are starting to steal application cycles from CPUs. Read more...

UCSD Researchers Identify Potential Bird Flu Drugs

Jul 02 | UC San Diego News Center | With the help of resources at the San Diego Supercomputer Center, UCSD scientists have isolated more than two dozen promising compounds from which new “designer drugs” might be developed to combat the avian flu virus. Read more...

Implementing Multi-Core: The Devil Is in the Detail

Jul 02 | Chip Design Magazine | Dual- and quad-core processors barely scratch the surface of the potential of multi-core systems. Read more...

Read more headlines...

Featured Whitepapers

New HPC White Paper: Star-P® Performance on IBM Linux Clusters

Jul 03 | | The paper explores some of the performance benefits of Star-P on commodity scalable systems such as IBM's Linux clusters based on multi-core Intel Xeon processors. The results demonstrate substantial performance gains with almost no programmer effort-roughly a 24-fold speed improvement for solving linear matrix equations. An overview of parallel computing with Star-P, a description of the performance test cases and description of IBM cluster configurations used for testing are also addressed.

Fast N-Body Simulation with CUDA C Compiler

Apr 17 | | An N-body simulation numerically approximates the evolution of a system of bodies in which each body continuously interacts with every other body, and it arises in many other computational science problems as well.

Improving Performance and Manageability for Seismic Processing and Imaging Applications with Parallel Storage

Jun 05 | | As pressure increases on the upstream seismic processing community to deliver ever-higher levels of productivity and efficiency, a new generation of storage solutions will be required that allow the maximum utilisation of high-performance computing (HPC) Linux cluster resources, together with the minimum of management overhead.

View the White Paper Library

Multimedia

Podcast: Interview with Ben Bennett of ClearSpeed Technology

Today, HPC organizations are requiring substantially more floating point performance to solve real-world problems. In this podcast, Ben Bennett, ClearSpeed General Manager, discusses how acceleration technology can improve the overall performance of standard x86-based systems...