BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//juliacon-2026//talk//NXU8WC
BEGIN:VTIMEZONE
TZID:CET
BEGIN:STANDARD
DTSTART:20001029T040000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:CET
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000326T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:CEST
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-juliacon-2026-NXU8WC@pretalx.com
DTSTART;TZID=CET:20260813T124500
DTEND;TZID=CET:20260813T130000
DESCRIPTION:We implemented a mixed-precision\, nested recursive Cholesky al
 gorithm for GPU Matrix Processing Units (NVIDIA H200\, AMD MI300X) using J
 ulia. With a hierarchical precision method\, we maximize throughput while 
 maintaining numerical stability. Our recursive SYRK achieves a 14x speedup
  over cuBLAS\, leading to a 5.32x overall speedup for Cholesky over cuSOLV
 ER FP64. The solver leverages Julia’s multiple dispatch to provide a por
 table interface for HPC.
DTSTAMP:20260502T104552Z
LOCATION:Room 6
SUMMARY:Hierarchical Precision and Recursion for Accelerating Symmetric Lin
 ear Solves on MXUs - Vicki Carrica
URL:https://pretalx.com/juliacon-2026/talk/NXU8WC/
END:VEVENT
END:VCALENDAR