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:20211207T054800Z
LOCATION:220-221
DTSTART;TZID=America/Chicago:20211116T153000
DTEND;TZID=America/Chicago:20211116T160000
UID:submissions.supercomputing.org_SC21_sess145_pap515@linklings.com
SUMMARY:Non-Recurring Engineering (NRE) Best Practices: A Case Study with 
 the NERSC/NVIDIA OpenMP Contract
DESCRIPTION:Paper\n\nNon-Recurring Engineering (NRE) Best Practices: A Cas
 e Study with the NERSC/NVIDIA OpenMP Contract\n\nDaley, Southwell, Gayatri
 , Biersdorff, Toepfer...\n\nThe NERSC supercomputer, Perlmutter, consists 
 of AMD CPUs and NVIDIA GPUs. NERSC users expect to be able to use OpenMP t
 o take advantage of the highly capable GPUs. This paper describes how NERS
 C/NVIDIA constructed a Non-Recurring Engineering (NRE) contract to add Ope
 nMP GPU-offload support to the NVIDIA HPC compilers. The paper describes h
 ow the contract incorporated the strengths of both parties and encouraged 
 collaboration to improve the quality of the final deliverable. We include 
 our best practices and how this particular contract took into account emer
 ging OpenMP specifications, NERSC workload requirements and how to use Ope
 nMP most efficiently on GPU hardware. This paper includes OpenMP applicati
 on performance results obtained with the NVIDIA compilers distributed in t
 he NVIDIA HPC SDK.\n\nTag: Reproducibility Badge, State of the Practice\n\
 nRegistration Category: Tech Program Reg Pass\n\nReproducibility Badges: A
 rtifact Available
END:VEVENT
END:VCALENDAR