Search for a command to run...
ATLAS is one of the two general-purpose detectors at the Large Hadron Collider (LHC). Its trigger system must reduce the anticipated proton collision rate of up to 40 MHz to a recordable event rate of 100-200 Hz. This is realized through a multi-level trigger system. The first-level trigger is implemented with custom-built electronics and makes an initial selection which reduces the rate to less than 100 kHz. The subsequent trigger selection is done in software run on PC farms. The first-level trigger decision is made by the central-trigger processor using information from coarse grained calorimeter information, dedicated muon-trigger detectors, and a variety of additional trigger inputs from detectors in the forward regions. We present the performance of the first-level trigger during the commissioning of the ATLAS detector during early LHC running. We cover the trigger strategies used during the different machine commissioning phases from first circulating beams and splash events to collisions. It is described how the very first proton events were successfully triggered using signals from scintillator trigger detectors in the forward region. For circulating and colliding beams electrostatic button pick-up detectors were used to clock the arriving proton bunches. These signals were immediately used to aid the timing in of the beams and the ATLAS detector. We describe the performance and timing in of the the first-level Calorimeter and muon trigger systems. The operation of the trigger relies on its real-time monitoring capabilities. We describe how trigger rates, timing information, and dead-time fractions were monitored to ensure the very good performance of the system.