Funded by the American Recovery and Reinvestment Act, the Advanced Networking Initiative, will ensure that the United States stays competitive in science and technology. Specifically, ESnet will develop a prototype 100 gigbits per second (Gbps) Ethernet network to connect DOE supercomputer centers at speeds 10 times faster than current technology. “This network will serve as a pilot for a future network-wide deployment of 100 Gbps Ethernet in research and commercial networks and represents a major step toward DOE’s vision of a 1-terabit—1,000 times faster than 1 gigabit—network interconnecting DOE Office of Science supercomputer centers” said Michael Strayer, head of DOE’s Office of Advanced Scientific Computing Research.
“ESnet has always been a service organization,” said Steve Cotter, ESnet Department Head at Berkeley Lab. “We exist to enable DOE scientists to do great work at the cutting edge, and to increase the scientific capabilities of the United States. The deployment of a next-generation 100 Gbps network will ensure that we continue to provide state-of-the-art services to our constituents and continue to enable scientific discovery.”
At a time when economic conditions are forcing private companies to cut back on investment in research and development, ESnet will be working with telecommunications companies and hardware vendors in bringing to market the latest networking technologies and deploying them in this pre-standards, prototype network.
As planned, some of the $62 million for this Initiative will be used both to create new jobs for network and software engineers at Berkeley Lab, but the bulk of the funding will be used for purchasing networking equipment or services from service providers who have the infrastructure to support the new 100 Gbps technology. In all, up to $59 million will be invested directly in the telecommunications industry in the United States.
In addition to the direct economic benefits of the project, there are induced ones as well. Several studies have shown that network investments provide both immediate and long-term benefits, ranging from higher wages to driving productivity and economic growth. By developing 100 Gbps now, more universities and companies will find 10 Gbps and 1 Gbps networks far more affordable.
A Science-Driven Need for More Bandwidth
DOE scientists are now generating data at the terabyte scale, and datasets will soon be in the petabyte range, or 1,000 terabytes. Moving this much data will require both greater bandwidth and reliability, as well as new protocols to enable these high-speed transfers.
The study of global climate change is a critical research area where the amount of data being created and accessed is growing exponentially. For example, an archive of past, present and future climate modeling data maintained by the Program for Climate Model Diagnosis and Intercomparison at Lawrence Livermore National Laboratory contains more than 35 terabytes of data and is accessed by more than 2,500 users worldwide. However, the next-generation archive is expected to contain at least 650 terabytes, and the larger distributed worldwide archive will be between 6 petabytes to 10 petabytes.
Another scientific driver for increased bandwidth is the Large Hadron Collider in Switzerland. Within this accelerator—the world’s largest—millions of protons racing at near the speed of light will collide every second; scientists suspect the outcome of these “subatomic smashups” will provide valuable insights into the origins of matter and dark energy in the Universe. The experiments will generate more data than the international scientific community has ever tried to manage—up to 100 gigabits per second, to be processed and analyzed by scientists around the globe.
National-scale Test Bed
As part of the Advanced Network Initiative’s approximately $59 million investment in new networking equipment and services, about $8 million to $9 million will go towards a national-scale network test bed for use by the research community and industry to test out new technologies, protocols and applications.
The test bed will consist of advanced network devices and components assembled to give network and middleware researchers the capabilities to prototype ESnet capabilities anticipated in the next decade. As host of the test bed, ESnet will develop strategies to move mature technologies from testing mode to production service.