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:20230831T095747Z
LOCATION:Schwarzhorn
DTSTART;TZID=Europe/Stockholm:20230628T150000
DTEND;TZID=Europe/Stockholm:20230628T153000
UID:submissions.pasc-conference.org_PASC23_sess160_msa138@linklings.com
SUMMARY:LLVM’s OpenMP Loop Transformations to Enhance Locality in OpenMP L
 oop Scheduling
DESCRIPTION:Minisymposium\n\nMichael Kruse (Argonne National Laboratory)\n
 \nThis talk will present OpenMP loop transformations, e.g., tiling, in a w
 ay that complements existing and novel dynamic schedules for OpenMP loops 
 and OpenMP tasking.  It can be used for auto-tuning for obtaining the opti
 mal composition of multiple loop transformations and parameters without th
 e cost of node-level dynamic scheduling, in particular to maximize spatial
  and temporal data locality. The loop transformations were first introduce
 d in OpenMP 5.1 and will be further developed in OpenMP 6.0. We also brief
 ly discuss iteration space reflection and mapping to different levels of t
 he hardware hierarchy of parallelism. The work is motivated by research ac
 tivities in the US DoE project PROTEAS-TUNE.\n\nDomain: Computer Science, 
 Machine Learning, and Applied Mathematics 
\n\nSession Chair: Vivek Kale (
 Sandia National Laboratories)
END:VEVENT
END:VCALENDAR