BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/New_York
X-LIC-LOCATION:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260422T143139Z
LOCATION:B310
DTSTART;TZID=America/New_York:20241118T110000
DTEND;TZID=America/New_York:20241118T113000
UID:submissions.supercomputing.org_SC24_sess754_ws_llvmf102@linklings.com
SUMMARY:Evaluating Tuning Opportunities of the LLVM/OpenMP Runtime
DESCRIPTION:Smeet Chheda, Gaurav Verma, Shilei Tian, and Barbara Chapman (
 Stony Brook University) and Johannes Doerfert (Lawrence Livermore National
  Laboratory (LLNL))\n\nTuning parallel applications on multi-core architec
 tures is an arduous task. Several studies have utilized auto-tuning for Op
 enMP applications via standardized user-facing features, namely number of 
 threads, thread placement and binding policy. In this paper, we analyze Op
 enMP application runtime through an exhaustive exploration of all relevant
  configuration options of the LLVM/OpenMP runtime.\n\nOur findings allow t
 o identify trends in tuning potential and architecture-aware tuning sugges
 tions. We will open-source the 240,000 unique samples collected during exp
 eriments. These runs have been conducted on three different CPU architectu
 res vital in the HPC and datacenter community. Choice of applications incl
 udes popular benchmark suites and microbenchmarks namely,  NPB, Barcelona 
 OpenMP Task Suite, XSBench, RSBench, SU3Bench and LULESH.\n\nWe employ Mac
 hine Learning algorithms to perform analysis, explain, and form qualitativ
 e relations between features comprising of the architecture, application, 
 input-size, threads, and environment variables.  This is further used to r
 ecommend different configurations given an application type/architecture.\
 n\nTag: Compilers, Parallel Programming Methods, Models, Languages and Env
 ironments, Performance Optimization\n\nRegistration Category: Workshop Reg
  Pass\n\nSession Chairs: James Brodman (AMD); Johannes de Fine Licht (Next
 Silicon Inc, ETH Zürich); Ryan Kabrick (Tactical Computing Laboratories LL
 C, University of Delaware); and Alexis Perry-Holby (Los Alamos National La
 boratory (LANL))\n\n
END:VEVENT
END:VCALENDAR