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:20230124T171525Z
LOCATION:C155
DTSTART;TZID=America/Chicago:20221113T091500
DTEND;TZID=America/Chicago:20221113T093700
UID:submissions.supercomputing.org_SC22_sess428_ws_p3hpc105@linklings.com
SUMMARY:Heterogeneous Programming for the Homogeneous Majority
DESCRIPTION:Workshop\n\nHeterogeneous Programming for the Homogeneous Majo
 rity\n\nDeakin, Cownie, Lin, McIntosh-Smith\n\nIn order to take advantage 
 of the burgeoning diversity in processors at the frontier of supercomputin
 g, the HPC community is migrating and improving codes to utilize heterogen
 eous nodes, where accelerators, principally GPUs, are highly prevalent in 
 top-tier supercomputer designs.  Programs therefore need to embrace at lea
 st some of the complexities of heterogeneous architectures.  Parallel prog
 ramming models have evolved to express heterogeneous paradigms whilst prov
 iding mechanisms for writing portable, performant programs.  History shows
  that technologies first introduced at the frontier percolate down to loca
 l workhorse systems. However, we expect there will always be a mix of syst
 ems, some heterogeneous, but some remaining as homogeneous CPU systems.  T
 hus it is important to ensure codes adapted for heterogeneous systems cont
 inue to run efficiently on CPUs.  In this study, we explore how well widel
 y used heterogeneous programming models perform on CPU-only platforms, and
  survey the performance portability they offer on the latest CPU architect
 ures.\n\nSession Format: Recorded\n\nTag: Performance Portability\n\nRegis
 tration Category: Workshop Reg Pass
END:VEVENT
END:VCALENDAR