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:20260522T150118Z
LOCATION:D173
DTSTART;TZID=America/Chicago:20181114T113000
DTEND;TZID=America/Chicago:20181114T120000
UID:submissions.supercomputing.org_SC18_sess487_exforum153@linklings.com
SUMMARY:High Performance OpenMP for GPUs
DESCRIPTION:Michael Wolfe (Nvidia Corporation, PGI Compilers and Tools)\n\
 nOpenMP has a 20 year history in HPC. It was the parallel programming mode
 l of choice for large SMP servers, and later proved effective for hybrid M
 PI+OpenMP programming on scalable systems.  As core counts rose, best prac
 tices for OpenMP implementations and programming have evolved. The early f
 ocus was on maximizing efficiency of loop-level parallelism.  More recentl
 y a global, outer-parallel approach has been used to improve scalability o
 n nodes with large numbers of CPU cores.  GPUs are optimized to deliver hi
 gh performance and throughput on massively parallel, regular, loop-structu
 red algorithms.  In this talk, we'll survey the features of the latest Ope
 nMP standards, describe how OpenMP programs can exploit the performance po
 tential of today's GPUs, and summarize the implications for a high-perform
 ance OpenMP implementation that supports both multicore CPUs and GPU accel
 erators.\n\nTag: Accelerators, Architectures, GPUs\n\nSession Chair: Xiao 
 Zhu (Purdue University)\n\n
END:VEVENT
END:VCALENDAR