BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Denver
X-LIC-LOCATION:America/Denver
BEGIN:DAYLIGHT
TZOFFSETFROM:-0700
TZOFFSETTO:-0600
TZNAME:MDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0600
TZOFFSETTO:-0700
TZNAME:MST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260422T000604Z
LOCATION:E Concourse
DTSTART;TZID=America/Denver:20231115T100000
DTEND;TZID=America/Denver:20231115T170000
UID:submissions.supercomputing.org_SC23_sess301_drs106@linklings.com
SUMMARY:Charged Particle Track Reconstruction Algorithms for Massively Par
 allel Systems
DESCRIPTION:Stephen Nicholas Swatman (University of Amsterdam, European Or
 ganization for Nuclear Research (CERN))\n\nThe reconstruction of the traje
 ctories of charged particles through detector experiments is a core comput
 ational task in the domain of high-energy physics. Upcoming upgrades to ac
 celerators such as the Large Hadron Collider as well as to experiments lik
 e ATLAS threaten to render existing CPU-based approaches to track reconstr
 uction insufficient, and the use of massively parallel systems - GPGPUs in
  particular - is an important opportunity to meet future data processing r
 equirements. In my thesis, I investigate the feasibility of GPGPU-based tr
 ack reconstruction from performance engineering perspective: I focus on st
 ructured analysis of application performance, the development of statistic
 al and analytical models of performance, methods for mitigating the challe
 nges of GPGPU programming, and the design and implementation of novel trac
 k reconstruction algorithms. The key contributions of my thesis include th
 e development of novel algorithms for hit clustering, seed finding, and co
 mbinatorial Kalman filtering, key parts of the track reconstruction proces
 s. These algorithms suffer from significant load imbalance and thread dive
 rgence, and I have developed a novel statistical method for estimating the
  performance effects of this, as well as to guide optimization through thr
 ead refinement and coarsening. I have developed a method for the automated
  design space exploration of data storage methods for magnetic fields, whi
 ch play a crucial role in track reconstruction. Furthermore, I have develo
 ped an evolutionary method for finding layouts for multi-dimensional array
 s in hierarchical memory systems. My thesis will be concluded by a compreh
 ensive study of the performance of track reconstruction, as guided by the 
 aforementioned research.\n\nTag: Accelerators, Artificial Intelligence/Mac
 hine Learning, Applications, Cloud Computing, Distributed Computing, Data 
 Analysis, Visualization, and Storage, Data Compression, Heterogeneous Comp
 uting, I/O and File Systems, Quantum Computing, Reproducibility, Security,
  Software Engineering\n\nRegistration Category: Tech Program Reg Pass, Exh
 ibits Reg Pass\n\n
END:VEVENT
END:VCALENDAR