BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/New_York
X-LIC-LOCATION:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20210402T160555Z
LOCATION:Track 8
DTSTART;TZID=America/New_York:20201112T172500
DTEND;TZID=America/New_York:20201112T175000
UID:submissions.supercomputing.org_SC20_sess214_ws_lasalss111@linklings.co
 m
SUMMARY:Implementation and Numerical Techniques for One Eflop/s HPL-AI Ben
 chmark on Fugaku
DESCRIPTION:Workshop\n\nImplementation and Numerical Techniques for One Ef
 lop/s HPL-AI Benchmark on Fugaku\n\nImamura, Kudo, Nitadori, Ina\n\nOur pe
 rformance benchmark of HPL-AI on the supercomputer Fugaku was awarded the 
 55th Top500. The effective performance was 1.42 EFlop/s, and the world's f
 irst achievement to exceed the wall of exascale in a floating-point arithm
 etic benchmark. Because HPL-AI is brand new and has no reference code for 
 large systems, several challenges in the large-scale benchmark emerge from
  a low-precision numerical viewpoint. It is not sufficient to replace FP64
  operations solely with those of FP32 or FP16. At the least, we need thoug
 htful numerical analysis for lower-precision arithmetic and the introducti
 on of optimization techniques on extensive computing such as on Fugaku. Th
 is study presents some technical analyses and insights on the accuracy iss
 ues, implementation and performance improvement, and report on the exascal
 e benchmark on Fugaku.\n\nTag: Algorithms, Extreme Scale Computing, Perfor
 mance/Productivity Measurement and Evaluation, Scalable Computing, Scienti
 fic Computing\n\nRegistration Category: Workshop Reg Pass
END:VEVENT
END:VCALENDAR