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:20230831T095746Z
LOCATION:Hall
DTSTART;TZID=Europe/Stockholm:20230627T193000
DTEND;TZID=Europe/Stockholm:20230627T213000
UID:submissions.pasc-conference.org_PASC23_sess116_pos134@linklings.com
SUMMARY:P01 - A Language-Interoperable C++-Based Memory-Manager for the IC
 ON Climate and Weather Prediction Model
DESCRIPTION:Poster\n\nClaudius Holeksa (Karlsruhe Institute of Technology)
 , Ralf Müller and Jörg Behrens (German Climate Computing Centre), Florian 
 Prill (DWD), Christopher Bignamini and Will Sawyer (ETH Zurich / CSCS), Xa
 vier Lapillonne (MeteoSwiss), Sergey Kosukhin and Daniel Klocke (Max Planc
 k Institute for Meteorology), Terry Cojean and Yen-Chen Chen (Karlsruhe In
 stitute of Technology), Hartwig Anzt (University of Tennessee), and Claudi
 a Frauen (German Climate Computing Centre)\n\nHPC machines now use acceler
 ators such as GPUs. In addition, CPUs themselves now feature many cores as
  well as special fast memory, like the Fujistu A64FX and Intel Sapphire Ra
 pids. These rapid changes create important challenges for simulation codes
  to accommodate different parallel programming models, but also orchestrat
 e processing units and memory locations. In this poster, we present a new 
 memory-manager concept for the ICON climate and weather prediction model t
 o store and manipulate model variables. This new memory-manager concept al
 lows going beyond OpenACC pragmas for GPU portability of the current Fortr
 an code, thanks to more fine-grained memory management. The memory manager
  is written in C++ which allows supporting vendor-native parallel programm
 ing frameworks which are also C++-based, like CUDA, HIP, and SYCL, or port
 ability layers like Kokkos. At the same time, the language-interoperabilit
 y enables keeping backward compatibility with Fortran while introducing co
 ncepts facilitating better ICON component interfaces, but also to graduall
 y and iteratively migrate parts of the code to newer languages for better 
 efficiency as needed. This memory-manager concept has been introduced with
 in the µphys subcomponent of ICON. In this poster, we show that beyond the
  ICON-specific aspects, this concept can be translated into many existing 
 simulation codes.
END:VEVENT
END:VCALENDAR