Data Logistics and the CMS Analysis Model

The Compact Muon Solenoid Experiment (CMS) at the Large HadronCollider (LHC) at CERN has brilliant prospects for uncovering newinformation about the physical structure of our universe. Soonphysicists around the world will participate together in analyzing CMSdata in search of new physics phenomena and the Higgs Boson. However,they face a significant problem: with 5 Petabytes of data needingdistribution each year, how will physicists get the data they need?How and where will they be able to analyze it? Computing resources andscientists are scattered around the world, while CMS data exists inlocalized chunks. The CMS computing model only allows analysis oflocally stored data, "tethering" analysis to storage. The VanderbiltCMS team is actively working to solve this problem with the Researchand Education Data Depot Network (REDDnet), a program run byVanderbilt's Advanced Computing Center for Research and Education(ACCRE). I participated in this effort by testing data transfers intoREDDnet via the gridFTP server, a File Transfer Protocol whichincorporates an LHC Computing Grid security layer. I created a testsuite which helped identify and solve a large number of problems withgridFTP. Once optimized, I achieved sustained throughputs of 700-800Megabits per second (Mbps) over a 1 Gigabit per second (Gbps)connection, with remarkably few failures. GridFTP is the gatewaybetween REDDnet and CMS, and my tests were designed to exercise andharden this important tool. My results support other indications thatthe REDDnet system will be a successful solution to the limitations ofdata-tethering in the CMS computing model.