BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Europe/Stockholm
X-LIC-LOCATION:Europe/Stockholm
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260522T162633Z
LOCATION:Bldg. 6 - Room 104
DTSTART;TZID=Europe/Stockholm:20260629T140000
DTEND;TZID=Europe/Stockholm:20260629T143000
UID:submissions.pasc-conference.org_PASC26_sess109_msa153@linklings.com
SUMMARY:You Never Really Cared About Double Precision, Did You?
DESCRIPTION:Alexandru Calotoiu, Pratyai Mazumder, and Torsten Hoefler (ETH
  Zurich)\n\nAs scientific applications, especially large Fortran simulatio
 ns, are increasingly expected to run efficiently on hardware originally de
 signed for machine learning workloads, performance portability has become 
 essential. At the same time, the reduced precision supported by such hardw
 are poses a major challenge. We argue that double-precision reference resu
 lts should no longer serve as the sole measure of correctness when evaluat
 ing code changes, whether semantic, syntactic, or introduced through manua
 l or automatic porting across architectures. Instead, we seek to discuss d
 ifferent levels of equivalence between codes that solve the same physical 
 problem, and how to better define tolerances and, ultimately, what constit
 utes a correct simulation result.\n\nDomain: Climate, Weather, and Earth S
 ciences\n\nSession Chair: Ahmad Nawab (ECMWF)\n\n
END:VEVENT
END:VCALENDAR