A Fast Local Algorithm for Track Reconstruction on Parallel Architectures

The reconstruction of particle trajectories, tracking, is a central process in the reconstruction of particle collisions in High Energy Physics detectors. At the LHCb detector in the Large Hadron Collider, bunches of particles collide 30 million times per second. These collisions produce about 10^9 particle trajectories per second that need to be reconstructed in real time, in order to filter and store data. Upcoming improvements in the LHCb detector will deprecate the hardware filter in favour of a full software filter, posing a computing challenge that requires a renovation of current algorithms and the underlying hardware. We present Search by triplet, a local tracking algorithm optimized for parallel architectures. We design our algorithm reducing Read-After-Write dependencies as well as conditional branches, incrementing the potential for parallelization. We analyze the complexity of our algorithm and validate our results. We show the scaling of our algorithm for an increasing number of collision events. We show sustained tests for our algorithm sequence given a simulated dataflow. We develop CPU and GPU implementations of our work, and hide the transmission times between device and host by executing a multi-stream pipeline. Our results provide a reliable basis for an informed assessment on the feasibility of LHCb event reconstruction on parallel architectures, enabling us to develop cost models for upcoming technology upgrades. The created software infrastructure is extensible and permits the addition of subsequent reconstruction algorithms.

[1]  Erik Lindholm,et al.  NVIDIA Tesla: A Unified Graphics and Computing Architecture , 2008, IEEE Micro.

[2]  M. Schiller,et al.  Track reconstruction and prompt $K^{0}_{S}$ production at the LHCb experiment , 2011 .

[3]  Peter Sanders,et al.  Parallel track reconstruction in CMS using the cellular automaton approach , 2014 .

[4]  David Rohr,et al.  GPU-accelerated track reconstruction in the ALICE High Level Trigger , 2017 .

[5]  C. Cheshkov,et al.  Fast Hough-transform track reconstruction for the ALICE TPC , 2006 .

[6]  Raul H. C. Lopes,et al.  A well-separated pairs decomposition algorithm for k-d trees implemented on multi-core architectures , 2014 .

[7]  Renato Quagliani,et al.  Status of HLT1 sequence and path towards 30 MHz , 2018 .

[8]  Olivier Callot,et al.  FastVelo, a fast and efficient pattern recognition package for the Velo , 2011 .

[9]  Gennady Ososkov,et al.  Filtering tracks in discrete detectors using a cellular automaton , 1993 .

[10]  Wen Chao Zhang,et al.  LHCb VELO Upgrade Technical Design Report , 2013 .

[11]  Francesco Caponio,et al.  The artificial retina processor for track reconstruction at the LHC crossing rate , 2014, 1409.1565.

[12]  Yuanjing Li,et al.  Track reconstruction based on Hough-transform for nTPC , 2014 .

[13]  Wen Chao Zhang,et al.  Framework TDR for the LHCb Upgrade : Technical Design Report , 2012 .

[14]  M. van Beuzekom.  Performance of the LHCb Vertex Locator , 2011, 2011 IEEE Nuclear Science Symposium Conference Record.

[15]  Vikas Singhal,et al.  Event selection for MUCH of CBM experiment using GPU computing , 2015, 2015 Annual IEEE India Conference (INDICON).

[16]  Omar Awile,et al.  An efficient low‐rank Kalman filter for modern SIMD architectures , 2018, Concurr. Comput. Pract. Exp..

[17]  D. Bruch,et al.  Online Data Reduction using Track and Vertex Reconstruction on GPUs for the Mu3e Experiment , 2017 .

[18]  G. Amdhal,et al.  Validity of the single processor approach to achieving large scale computing capabilities , 1967, AFIPS '67 (Spring).

[19]  T. Başar,et al.  A New Approach to Linear Filtering and Prediction Problems , 2001 .

[20]  M. Altarelli,et al.  LHCb: detector performance and first physics results , 2011, 1105.5330.