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:20230831T095745Z
LOCATION:Sanada I
DTSTART;TZID=Europe/Stockholm:20230626T143000
DTEND;TZID=Europe/Stockholm:20230626T150000
UID:submissions.pasc-conference.org_PASC23_sess171_msa172@linklings.com
SUMMARY:Energy Impact of Kernel Level Optimization in ICON
DESCRIPTION:Minisymposium\n\nRamona Hohl, Benedikt Dorschner, Dmitry Alexe
 ev, and Peter Messmer (NVIDIA Inc.)\n\nGeneral code complexity, together w
 ith portability and maintainability requirements make weather and climate 
 codes challenging to optimize. Directive-based approaches such as OpenACC 
 or domain specific languages are therefore preferred choices to achieve hi
 gh performance while retaining portability and developer productivity. How
 ever, the increasing complexity of modern hardware with deep memory hierar
 chies, multiple levels of parallelism and varieties of complex functional 
 units make it harder for the compiler to generate optimal code. Comparing 
 performance of the generated code to hand tuned implementations is therefo
 re necessary. Such comparisons do not only demonstrate the performance tha
 t is left on the table, but also the minimum energy to solution, an increa
 singly important metric in today’s energy constrained environment. Typical
 ly, such hand-crafted implementations are limited to relatively small kern
 els, leading to missed optimization opportunities across a larger scope. I
 n this presentation, we will therefore discuss a CUDA port of a larger com
 ponent of ICON and analyze the impact of these optimization on the overall
  energy footprint. We will discuss the practical aspects of collecting ene
 rgy measurements on GPUs and explore the impact of hardware features, whic
 h are challenging to access in a directive-based environment.\n\nDomain: C
 limate, Weather and Earth Sciences\n\nSession Chair: Peter Messmer (NVIDIA
  Inc.)
END:VEVENT
END:VCALENDAR