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:20220812T074334Z
LOCATION:Foyer 2nd Floor
DTSTART;TZID=Europe/Stockholm:20220628T090000
DTEND;TZID=Europe/Stockholm:20220628T110000
UID:submissions.pasc-conference.org_PASC22_sess181_pos153@linklings.com
SUMMARY:P50 - Recent Atlas Library Developments for Earth System Modelling
DESCRIPTION:Poster\n\nP50 - Recent Atlas Library Developments for Earth Sy
 stem Modelling\n\nDeconinck\n\nECMWF has the strategy to run ECMWF's Integ
 rated Forecasting System (IFS) -- or at least some of its Earth sytem mode
 l components -- on non-traditional hardware such as GPUs. <br />In this st
 rategy, the ECMWF Atlas library plays a central role to manage data struct
 ures, distributed parallelisation, and memory spaces. The Atlas library po
 ssibly delegates its field data allocations which contains links between h
 ost CPU memory spaces and GPU device memory spaces to GridTools, developed
  at CSCS/ETH/MeteoSwiss. GridTools is a domain specific language (DSL) for
  stencil based numerical algorithms, and Atlas can then encapsulate the fi
 elds used in such numerical algorithm. Orthogonally, IFS consists of vario
 us Earth system model components such as atmosphere, radiation, ocean, oce
 an waves. These components may each operate on different grids and use dif
 ferent parallel distributions. Adding GPU device memory spaces to this mix
  is inevitably making the coupling of these components more challenging. E
 CMWF therefore aims to integrate Atlas deeper in each Earth system model c
 omponent, and thus take advantage of a common datastructure library to eff
 iciently interpolate fields between each component. The presented poster w
 ill elaborate on the recent Atlas developments to bring us closer to our a
 mbitious goals.
END:VEVENT
END:VCALENDAR
