BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Chicago
X-LIC-LOCATION:America/Chicago
BEGIN:DAYLIGHT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
TZNAME:CDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
TZNAME:CST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260202T201809Z
LOCATION:263-264
DTSTART;TZID=America/Chicago:20251118T105200
DTEND;TZID=America/Chicago:20251118T111500
UID:submissions.supercomputing.org_SC25_sess184_pap769@linklings.com
SUMMARY:Scaling the Memory Wall Using Mixed-Precision — HPG-MxP on an Exas
 cale Machine
DESCRIPTION:Aditya Kashi, Nicholson Koukpaizan, Hao Lu, Michael Matheson, 
 Sarp Oral, and Feiyi Wang (Oak Ridge National Laboratory (ORNL))\n\nMixed-
 precision algorithms have been proposed as a way for scientific computing 
 to benefit from some of the gains seen for AI on recent high performance c
 omputing (HPC) platforms. A few applications dominated by dense matrix ope
 rations have seen substantial speedups by utilizing low-precision formats 
 such as FP16. However, a majority of scientific simulation applications ar
 e memory bandwidth limited. Beyond preliminary studies, the practical gain
  from using mixed-precision algorithms on a given HPC system is largely un
 clear.\n\nThe High Performance GMRES Mixed Precision (HPG-MxP) benchmark h
 as been proposed to measure the useful performance of an HPC system on spa
 rse matrix-based mixed-precision applications. In this work, we present a 
 highly optimized implementation of the HPG-MxP benchmark for an exascale s
 ystem and describe our algorithm enhancements. We show for the first time 
 a speedup of 1.6x using a combination of double and single precision on mo
 dern GPU-based supercomputers.\n\nTag: Performance Measurement, Modeling, 
 & Tools\n\nRecording: Livestreamed, Recorded\n\nRegistration Category: Tec
 hnical Program Reg Pass\n\nSession Chair: Ian Karlin (NVIDIA Corporation)\
 n\n
END:VEVENT
END:VCALENDAR