CMS Computing: Performance and Outlook

After years of development, the CMS distributed computing system is now in full operation. The LHC continues to set records for instantaneous luminosity, and CMS continues to record data at 300 Hz. Because of the intensity of the beams, there are multiple proton-proton interactions per beam crossing, leading to larger and larger event sizes and processing times. The CMS computing system has responded admirably to these challenges. We present the current status of the system, describe the recent performance, and discuss the challenges ahead and how we intend to meet them. 1. The problem Experiments at the Large Hadron Collider (LHC) [1] will produce tremendous amounts of data. With instantaneous luminosities of 1034 cm−2s−1 and a crossing rate of 40 MHz, the collision rate will be about 109 Hz. But the rate for new physics processes, after accounting for branching fractions and the like, is of order 10−5 Hz, leading to the need to select events out of a huge data sample at the level of 10−14. The Compact Muon Solenoid (CMS) experiment [2] developed a distributed computing model from the very early days of the experiment. There are a variety of motivating factors for this: a single data center at CERN would be expensive to build and operate, whereas smaller data centers at multiple sites are less expensive and can leverage local resources (both financial and human). But there are also many challenges in making