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:20260202T201805Z
LOCATION:121
DTSTART;TZID=America/Chicago:20251116T083000
DTEND;TZID=America/Chicago:20251116T120000
UID:submissions.supercomputing.org_SC25_sess256_tut122@linklings.com
SUMMARY:Tools To Detect and Diagnose Floating-Point Errors in Heterogeneou
 s Computing Hardware and Software
DESCRIPTION:Ganesh Gopalakrishnan (University of Utah); Ignacio Laguna (La
 wrence Livermore National Laboratory (LLNL)); and Cindy Rubio-González (Un
 iversity of California, Davis)\n\nHigh performance computing and machine l
 earning applications increasingly rely on mixed-precision arithmetic on CP
 Us and GPUs for superior performance. However, this shift introduces sever
 al challenging numerical issues such as increased round-off errors, and IN
 F and NaN exceptions that can render the computed solutions useless. At pr
 esent, this places a heavy burden on developers, interrupting their work w
 hile they diagnose these problems manually. This tutorial presents three t
 ools that target specific issues leading to floating-point bugs. First, we
  present FPChecker, which not only detects and reports INF/NaN exceptions 
 in parallel and distributed CPU codes, but also tells programmers about th
 e exponent value ranges for avoiding exceptions while also minimizing roun
 ding errors. Second, we present GPU-FPX, which detects floating-point exce
 ptions generated by NVIDIA GPUs, including their Tensor Cores via a "nixna
 n" extension to GPU-FPX. Third, we present FloatGuard, a unique tool that 
 detects exceptions in AMD GPUs. The tutorial is aimed at helping programme
 rs avoid exception bugs; for this, we will demonstrate our tools on simple
  examples with seeded bugs. Attendees may optionally install and run our t
 ools. The tutorial also allocates question/answer time to address real sit
 uations faced by the attendees.\n\nRecording: Livestreamed, Recorded\n\nRe
 gistration Category: Tutorial Reg Pass\n\n
END:VEVENT
END:VCALENDAR