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:20210916T132451Z
LOCATION:Jean Calvin
DTSTART;TZID=Europe/Stockholm:20210707T140000
DTEND;TZID=Europe/Stockholm:20210707T143000
UID:submissions.pasc-conference.org_PASC21_sess148_msa146@linklings.com
SUMMARY:ESCAPE 2: Energy-efficient SCalable Algorithms for Weather and Cli
 mate Prediction at Exascale
DESCRIPTION:Minisymposium\n\nESCAPE 2: Energy-efficient SCalable Algorithm
 s for Weather and Climate Prediction at Exascale\n\nMueller\n\nIn the simu
 lation of complex multi-scale flow problems, such as those arising in weat
 her and climate modelling, one of the biggest challenges is to satisfy ope
 rational requirements in terms of time-to-solution and available energy wi
 thout compromising the accuracy and stability of the solution. These compe
 ting factors require extreme computational capabilities in conjunction wit
 h state-of-the-art algorithms that can optimally suit the targeted underly
 ing hardware while improving the convergence to the desired solution. The 
 European Centre for Medium Range Weather Forecasts (ECMWF) is leading the 
 ESCAPE projects funded by Horizon 2020 under initiative Future and Emergin
 g Technologies in High Performance Computing. The participating models are
  broken down into smaller building blocks called dwarfs. These are then op
 timised for different hardware architectures and alternative algorithms ar
 e investigated. It was shown that GPU optimisation which takes full advant
 age of NVLink interconnect can provide a massive speedup (23x for spectral
  transform and 57x for MPDATA) if all computations are run on the GPU. Thi
 s work enabled ECMWF to make good use of the GPUs of the Summit supercompu
 ter. This talk will give an overview of the optimisations required to achi
 eve good performance on Summit and the results obtained.\n\nDomain: CS and
  Math, Climate and Weather
END:VEVENT
END:VCALENDAR
