Implementation and Performance of the Alignment of the ATLAS Inner Detector Tracking System

The Large Hadron Collider (LHC) at CERN is the world's largest particle accelerator. After a successful initial run at 900 GeV in 2009, during 2010, LHC has collided proton beams at an unprecedented centre of mass energy of 7 TeV. ATLAS is a multipurpose experiment that records the products of the LHC collisions. To reconstruct trajectories of charged particles produced in these collisions, ATLAS is equipped with a tracking system built on two different technologies: silicon planar sensors and drift-tube based detectors, constituting the ATLAS Inner Detector. In order to achieve its scientific goals, ATLAS tracking performance requirements are quite exigent. The goal of the alignment is set such that the limited knowledge of the sensor locations should not deteriorate the resolution of the track parameters by more than 20% with respect to the intrinsic tracker resolution. Thus the required precision for the alignment of the silicon sensors is below 10 micrometers. The alignment of the ATLAS tracking system requires the determination of its almost 36000 degrees of freedom (DoF) with high accuracy. This demands to use a large sample of high momentum and isolated charge particle tracks. The new implementation of the track based alignment within the ATLAS software framework unifies different alignment approaches and allows alignment of all tracking subsystems together. The alignment specific classes are directly linked with the tra ck reconstruction software, which provides tools for computation of specific quantities needed for the alignment (residuals, pulls, derivatives, ...). The detector specific classes allow a common definition of the alignment geometry. As the alignment algorithms are based on minimization of the track-hit residuals that leads to solving of linear system with large number of DoF. The solving itself poses a real challenge as it involves inversion or diagonalization of a large matrix that may be dense. Inversion with fast solver algorithms as well as full diagonalization have been carried out producing results for the alignment of the real detector. The alignment jobs are executed at the CERN Analysis Facility with ~200 CPUs running Sci! entific Linux CERN 5. It is also possible to run the alignment jobs on the GRID. In both cases the event processing is run in parallel in many jobs. The output matrices from all jobs are added before solving. We will present the outline of the track based alignment approaches, their implementation within the ATLAS software framework and performance when aligning the ATLAS detector using real data. The alignment has been performed on data recorded from both: the LHC collisions and cosmic ray data. Results obtained with 2009 and 2010 data will be shown.