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:20260422T000711Z
LOCATION:607
DTSTART;TZID=America/Denver:20231112T103000
DTEND;TZID=America/Denver:20231112T105000
UID:submissions.supercomputing.org_SC23_sess415_ws_esp104@linklings.com
SUMMARY:Uncertainty Quantification of Reduced-Precision Time Series in Tur
 bulent Channel Flow
DESCRIPTION:Martin Karp (KTH Royal Institute of Technology, Sweden); Felix
  Liu (KTH Royal Institute of Technology, Sweden; Raysearch Laboratories); 
 Ronith Stanly (KTH Royal Institute of Technology, Sweden); Saleh Rezaeirav
 esh (University of Manchester); Niclas Jansson (KTH Royal Institute of Tec
 hnology, Sweden); Philipp Schlatter (Friedrich-Alexander University, Erlan
 gen-Nuremberg; KTH Royal Institute of Technology, Sweden); and Stefano Mar
 kidis (KTH Royal Institute of Technology, Sweden)\n\nWith increased comput
 ational power through the use of low-precision arithmetic, a relevant ques
 tion is how lower precision affects simulation results, especially for cha
 otic systems where analytical round-off estimates are non-trivial to obtai
 n. In this work, we consider how the uncertainty of the time series of a d
 irect numerical simulation of turbulent channel flow at 𝑅𝑒𝜏 = 180 is affec
 ted when restricted to a reduced-precision representation. We utilize a no
 n-overlapping batch means estimator and find that the mean statistics can,
  in this case, be obtained with significantly fewer mantissa bits than con
 ventional IEEE-754 double precision, but that the mean flow is more sensit
 ive in the middle of the channel than the boundary layer. This indicates t
 hat using lower precision in the boundary layer, where the majority of the
  computational work is located, may benefit significantly from low-precisi
 on floating point units found in upcoming computer hardware.\n\nTag: Perfo
 rmance Optimization\n\nRegistration Category: Workshop Reg Pass\n\nSession
  Chairs: Tiernan Casey (Sandia National Laboratories) and Antigoni Georgia
 dou (Oak Ridge National Laboratory (ORNL))\n\n
END:VEVENT
END:VCALENDAR