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:20230124T171524Z
LOCATION:C155
DTSTART;TZID=America/Chicago:20221114T110000
DTEND;TZID=America/Chicago:20221114T113000
UID:submissions.supercomputing.org_SC22_sess453_ws_pmbsf113@linklings.com
SUMMARY:Evaluating ISO C++ Parallel Algorithms on Heterogeneous HPC System
 s
DESCRIPTION:Workshop\n\nEvaluating ISO C++ Parallel Algorithms on Heteroge
 neous HPC Systems\n\nLin, Deakin, McIntosh-Smith\n\nRecent revisions to th
 e ISO C++ standard have added specifications for parallel algorithms.  The
 se additions cover common use-cases, including sequence traversal, reducti
 on, and even sorting, many of which are highly applicable in HPC, and thus
  represent a potential for increased performance and productivity.\n\nThis
  study evaluates the state of the art for implementing heterogeneous HPC a
 pplications using the latest built-in ISO C++17 parallel algorithms.  We i
 mplement C++17 ports of representative HPC mini-apps that cover both compu
 te-bound and memory bandwidth-bound applications.  We then conduct benchma
 rks on CPUs and GPUs, comparing our ports to other widely-available parall
 el programming models, such as OpenMP, CUDA, and SYCL.\n\nFinally, we show
  that C++17 parallel algorithms are able to achieve competitive performanc
 e across multiple mini-apps on many platforms, with some notable exception
 s.  We also discuss several key topics, including portability, and describ
 e workarounds for a number of remaining issues, including index-based trav
 ersal and accelerator device/memory management.\n\nSession Format: Recorde
 d\n\nTag: Applications, Architectures, Benchmarking, Exascale Computing, M
 odeling and Simulation, Performance, Performance Portability\n\nRegistrati
 on Category: Workshop Reg Pass
END:VEVENT
END:VCALENDAR