The ATLAS Fast TracKer—Architecture, Status and High-Level Data Quality Monitoring Framework
The Fast Tracker (FTK) is a highly parallel processor dedicated to a quick and efficient reconstruction of tracks in the Pixel and Semiconductor Tracker (SCT) detectors of the ATLAS experiment at LHC. It is designed to identify charged particle tracks with transverse momentum above 1 GeV and reconstruct their parameters at an event rate of up to 100 kHz. The average latency of the processing is below 100 μs at the expected collision intensities. This performance is achieved by using custom ASIC chips with associative memory for pattern matching, while modern FPGAs calculate the track parameters. This paper describes the architecture, the current status and a High-Level Data Quality Monitoring framework of the FTK system. This monitoring framework provides an online comparison of the FTK hardware output with the FTK functional simulation, which is run on the pixel and SCT detector data at a low rate, allowing the detection of non-expected outputs of the FTK system.