USLHCNet provides transatlantic connections of the Tier1 computing facilities at Fermilab and Brookhaven with the Tier0 and Tier1 facilities at CERN as well as Tier1s elsewhere in Europe and Asia. Together with [ESnet], Internet2, the GEANT pan-European network, and NSF’s UltraLight project, US LHCNet also supports connections between the Tier2 centers (where most of the analysis of the data will take place, starting this year) and the Tier1s as needed.
USLHCNet today consists of a set of 10 Gbps links interconnecting CERN, MANLAN in New York and Starlight in Chicago. The network has been architected to ensure efficient and reliable use of the 10 Gbps bandwidth of each link, up to relatively high occupancy levels, to cover a wide variety of network tasks, including: large file transfers, grid applications, data analysis sessions involving client-server software as well as simple remote login, network and grid R&D-related traffic, videoconferencing, and general Internet connectivity. Caltech shares with CERN the responsibility for the implementation, operation and management of the US-CERN network, as well as the peerings with ESnet and the major US academic network backbones (in particular Abilene and more recently National Lambda Rail). Funding for the network bandwidth, the routing and switching equipment and other infrastructure required for network operation, is shared by Caltech (under a DOE/HEP grant) and CERN.
The effectiveness of US participation in the LHC experimental program is particularly dependent on the speed and reliability of our national and international networks. As we approach the startup of the LHC, the ability of scientists to move large amounts of data, access computing and data resources, and collaborate in real time from multiple remote locations require unprecedented network performance and reliability.
In addition to the application of state of the art high-throughput methods and tools, US LHCNet has been designed to meet these needs by providing a high performance network with 99.9+% availability, through the use of multiple links across the Atlantic, network equipment that provides robust fallback at the optical layer in case of link failure, and automatic re-direction of network traffic using redundant network equipment at each of the US LHCNet points of presence (PoPs).