BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Europe/Stockholm
X-LIC-LOCATION:Europe/Stockholm
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20190719T085743Z
LOCATION:HG D 1.1
DTSTART;TZID=Europe/Stockholm:20190612T160000
DTEND;TZID=Europe/Stockholm:20190612T163000
UID:submissions.pasc-conference.org_PASC19_sess147_msa203@linklings.com
SUMMARY:Enabling ICON for Kilometer-Scale Global Climate on GPU Systems
DESCRIPTION:Minisymposium\nClimate and Weather\n\nEnabling ICON for Kilome
 ter-Scale Global Climate on GPU Systems\n\nSawyer\n\nThe Icosahedral Non-h
 ydrostatic (ICON) model employs a finite-volume solver of the equations of
  atmospheric motion and the physical parameterizations from the well-known
  ECHAM model. It is being ported to accelerators within the PASC-funded EN
 IAC project, utilizing OpenACC accelerator directives as well as the CLAW 
 source-to-source translator, which generates OpenACC or OpenMP directives 
 from CLAW meta-directives.  This approach is intended as an intermedi
 ate solution to allow single-source compatibility with the existing Fortra
 n code base, while developers slowly embrace emerging paradigms such as do
 main specific languages (DSLs).  For a memory-bandwidth-limited appli
 cation such as ICON, this technique can achieve good performance as long a
 s the GPU is sufficiently occupied.  With ICON's near perfect weak sc
 aling, this implies that a GPU platform like Piz Daint at the Swiss Nation
 al Supercomputing Centre (CSCS) with 5272 nodes can realistically simulate
  kilometer-scale global climate, e.g., to model the Quasi-Biennial Oscilla
 tion (QBO). We present first simulations with ICON at high resolution on l
 arge GPU configurations, draw some conclusions about the efficacy of accel
 erators for atmospheric models, and discuss the productivity benefit of th
 is bridging technology as we progress to Exascale.
END:VEVENT
END:VCALENDAR