Is a world-scale Lambda-based Laboratory for application and middleware development on emerging LambdaGrids, where applications rely on dynamically configured networks based on optical wavelengths. GLIF was established at the third annual LambdaGrid Workshop organized by Kees Neggers of SURFnet and Cees de Laat of the University of Amsterdam, and hosted by NORDUnet at their annual conference in Reykjavik, Iceland, in August 2003.

The outcome of this workshop was the need to move from one-time demonstrations to support persistent application development. Since the LambdaGrid is growing and reaching more and more research groups, which are already collaborating on the application level, this should become easier.

The GLIF community shares a common vision of building a new grid-computing paradigm, in which the central architectural element is optical networks, not computers, to support this decade's most demanding e-science applications. This paradigm is caused by the use of parallelism, as in supercomputing a decade ago; however, the parallelism is in multiple wavelengths of light, or lambdas, on single optical fibers, creating "supernetworks."

Today's Grid is built on these "best effort" shared TCP/IP networks. In other words, the network is simply the glue that holds the middleware-enabled computational resources together. In contrast, we are interested in developing "application-empowered" networks, in which the networks themselves are schedulable Grid resources. These application-empowered deterministic networks, or "LambdaGrids," complement the conventional "best effort" networks that provide a general infrastructure with a common set of services to the broader research and education community. They are becoming a necessary component of the USA cyberinfrastructure, European e-infrastructure and proposed Canadian i-infrastructure initiatives.

The main application drivers for these new "application-empowered" networks are high-performance e-science projects, where e-science represents very large-scale applications - such as high-energy physics, astronomy, earth science, bioinformatics, environmental - that study very complex micro to macro-scale problems over time and space. In the future, these networks will conceivably migrate to other domains, including education, emergency services, health services and commerce. E-science will require distributed petaops computing, exabyte storage, and terabit networks in the coming decade.

What is Lambda Networking and a LambdaGrid?

Lambda-based networking is ultimately about using different "colors" or wavelengths of (laser) light in fibers for separate connections. Each wavelength is called a "lambda." Current coding schemes allow for typically

10 Gbps to be encoded by a laser on a high-speed network interface. In lambda networking, the goal is to achieve ultimate Quality of Service by giving applications and user communities their own sets of lambdas on a shared (dark) fiber infrastructure; thus, isolating the different communities from each other. The implementation requires Dense Wavelength Division Multiplexing (DWDM) to accommodate many wavelengths on a fiber, optical switches (e.g., based on MEMS), and other optical networking equipment. A LambdaGrid requires the interconnectivity of optical links, each carrying one or more lambdas, or wavelengths, of data, to form on-demand, end-to-end "light paths," in order to meet the needs of very demanding e-science applications.

GLIF History?

In 2001, SURFnet and TERENA organized and hosted the first invitation-only LambdaGrid meeting, followed by an open LambdaGrid Workshop. The first research-only 2.5Gbps lambda between NetherLight and StarLight was on order.

At that meeting, concepts about lambda networking were established and the first experiments were defined.

In 2002, iGrid 2002 was an open event, followed by the second, invitation-only LambdaGrid Workshop, this time hosted by the Science Park Amsterdam and SURFnet. At that meeting, lessons learned from the first year and the iGrid 2002 event and the expansion of the testbed were discussed.