Kalman filter tracking on parallel architectures

Cerati, G. B.; Elmer, P.; Krutelyov, S.; Lantz, Steven; Lefèbvre, Matthieu; Mcdermott, K.; Riley, D.; Tadel, M.; Wittich, P.; Würthwein, F.; Yagil, Avi

doi:10.1088/1742-6596/898/4/042051

Cited by 9 publications

(8 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Beyond 32 threads the standard work chunk of 32 seeds per task gets progressively reduced down to 4 seeds per task at 256 threads. See [7] for details about how multithreading is implemented in mkFit. SKL-SP shows good scaling up to 8 threads and KNL up to 32 threads.…”

Section: Single Event Performance Of Core Track Findingmentioning

confidence: 99%

See 1 more Smart Citation

Parallelized and Vectorized Tracking Using Kalman Filters with CMS Detector Geometry and Events

et al. 2019

View full text Add to dashboard Cite

The High-Luminosity Large Hadron Collider at CERN will be characterized by greater pileup of events and higher occupancy, making the track reconstruction even more computationally demanding. Existing algorithms at the LHC are based on Kalman filter techniques with proven excellent physics performance under a variety of conditions. Starting in 2014, we have been developing Kalman-filter-based methods for track finding and fitting adapted for many-core SIMD processors that are becoming dominant in high-performance systems. This paper summarizes the latest extensions to our software that allow it to run on the realistic CMS-2017 tracker geometry using CMSSW-generated events, including pileup. The reconstructed tracks can be validated against either the CMSSW simulation that generated the detector hits, or the CMSSW reconstruction of the tracks. In general, the code's computational performance has continued to improve while the above capabilities were being added. We demonstrate that the present Kalman filter implementation is able to reconstruct events with comparable physics performance to CMSSW, while providing generally better computational performance. Further plans for advancing the software are

show abstract

Section: Single Event Performance Of Core Track Findingmentioning

confidence: 99%

“…OpenMP was replaced by Intel Thread Building Blocks (TBB) to increase flexibility as well as to be in compliance with the CMS code base [7]. Further, to avoid imbalances in η regions and to provide more workload tasks for the many available cores, support for processing of multiple events in parallel was added.…”

Section: Introductionmentioning

confidence: 99%

Parallelized and Vectorized Tracking Using Kalman Filters with CMS Detector Geometry and Events

et al. 2019

View full text Add to dashboard Cite

show abstract

“…Relying on this library, the initial implementation of a vectorized KF track fitting algorithm was demonstrated in a simplified detector geometry [4], followed by the initial implementation of an analogous track finding algorithm [5]. Further developments [6][7][8] allowed the software to achieve satisfactory performance, using Intel Threading Building Blocks (TBB) parallel constructs to comply with the CMS software (CMSSW) code base. A first implementation of the mkFit program on GPGPUs was also pursued [9]: while the Matriplex library was found to outperform standard small-matrix multiplication packages for GPUs, the performance of the KF-based track finding on GPGPUs is as of today not satisfactory.…”

Section: Parallelized Kalman Filter Trackingmentioning

confidence: 99%

Parallelized Kalman-Filter-Based Reconstruction of Particle Tracks on Many-Core Architectures

Cerati

Elmer

Krutelyov

et al. 2018

J. Phys.: Conf. Ser.

Self Cite

View full text Add to dashboard Cite

In the High-Luminosity Large Hadron Collider (HL-LHC), one of the most challenging computational problems is expected to be finding and fitting charged-particle tracks during event reconstruction. The methods currently in use at the LHC are based on the Kalman filter. Such methods have shown to be robust and to provide good physics performance, both in the trigger and offline. In order to improve computational performance, we explored Kalman-filter-based methods for track finding and fitting, adapted for many-core SIMD and SIMT architectures. Our adapted Kalman-filter-based software has obtained significant parallel speedups using such processors, e.g., Intel Xeon Phi, Intel Xeon SP (Scalable Processors) and (to a limited degree) NVIDIA GPUs. Recently, an effort has started towards the integration of our software into the CMS software framework, in view of its exploitation for the Run III of the LHC. Prior reports have shown that our software allows in fact for some significant improvements over the existing framework in terms of computational performance with comparable physics performance, even when applied to realistic detector configurations and event complexity. Here, we demonstrate that in such conditions physics performance can be further improved with respect to our prior reports, while retaining the improvements in computational performance, by making use of the knowledge of the detector and its geometry.

show abstract

“…As the hit collection is completely determined after track building, track fitting can repeatedly apply the KF algorithm without branching, making this the ideal place to start in porting KF tracking to Xeon and Xeon Phi, with our first results shown at ACAT2014 [8]. In more recent works (CHEP2016 [11], CtD2016 [12]), we discussed the improvements made to target a more realistic geometry. Profiling and recent computational optimizations were also outlined.…”

Section: Kalman Filter Trackingmentioning

confidence: 99%

“…This is in contrast to a first attempt at the combinatorial KF which copied a candidate each time a hit was deemed compatible, then sorting and keeping only the best N candidates per seed after all the possible hits on a given layer for all input candidates were explored. A more detailed discussion of this work on memory management and impacts on performance was presented at NSS-MIC2015 [10] and at CHEP 2016 [11]…”

Section: Memory Managementmentioning

confidence: 99%

Parallelized Kalman-Filter-Based Reconstruction of Particle Tracks on Many-Core Processors and GPUs

et al. 2017

View full text Add to dashboard Cite

Abstract. For over a decade now, physical and energy constraints have limited clock speed improvements in commodity microprocessors. Instead, chipmakers have been pushed into producing lower-power, multi-core processors such as Graphical Processing Units (GPU), ARM CPUs, and Intel MICs. Broad-based efforts from manufacturers and developers have been devoted to making these processors user-friendly enough to perform general computations. However, extracting performance from a larger number of cores, as well as specialized vector or SIMD units, requires special care in algorithm design and code optimization. One of the most computationally challenging problems in high-energy particle experiments is finding and fitting the charged-particle tracks during event reconstruction. This is expected to become by far the dominant problem at the High-Luminosity Large Hadron Collider (HL-LHC), for example. Today the most common track finding methods are those based on the Kalman filter. Experience with Kalman techniques on real tracking detector systems has shown that they are robust and provide high physics performance. This is why they are currently in use at the LHC, both in the trigger and offline. Previously we reported on the significant parallel speedups that resulted from our investigations to adapt Kalman filters to track fitting and track building on Intel Xeon and Xeon Phi. Here, we discuss our progresses toward the understanding of these processors and the new developments to port the Kalman filter to NVIDIA GPUs. a

show abstract

Kalman filter tracking on parallel architectures

Cited by 9 publications

References 4 publications

Parallelized and Vectorized Tracking Using Kalman Filters with CMS Detector Geometry and Events

Parallelized and Vectorized Tracking Using Kalman Filters with CMS Detector Geometry and Events

Parallelized Kalman-Filter-Based Reconstruction of Particle Tracks on Many-Core Architectures

Parallelized Kalman-Filter-Based Reconstruction of Particle Tracks on Many-Core Processors and GPUs

Contact Info

Product

Resources

About