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:Darwin Room
DTSTART;TZID=Europe/Stockholm:20220627T140000
DTEND;TZID=Europe/Stockholm:20220627T143000
UID:submissions.pasc-conference.org_PASC22_sess119_msa139@linklings.com
SUMMARY:Alpaka, LLAMA and other Animals - Making PIConGPU Ready for Exasca
 le
DESCRIPTION:Minisymposium\n\nAlpaka, LLAMA and other Animals - Making PICo
 nGPU Ready for Exascale\n\nBussmann, Debus, Kluge, Widera, Steiniger...\n\
 nThe quest for the best architetcure in HPC is ongoing. While the transiti
 on from CPUs to GPU architectures has already happened, new architetcures 
 with more intelligence in chip and better control of workflows are on the 
 horizon. Particle-in-cell codes are notoriously hard to optimize for highl
 y parallel hardware. Challenges include porting parallelized parts of the 
 code to new coding paradigms every few years, optimization of data layouts
  for optimum memory throughput, increasing utilization of parallelism by u
 nderstanding data dependencies and, last but not least, enabling fast I/O 
 and complex AI/HPC/Visualization workflows.<br /><br />In this talk we wil
 l present our solutions ALPAKA, LLAMA, REDGRAPES, openPMD and ISAAC to sho
 wcase the ecosystem from which our Exascale-ready particle-in-cell code PI
 ConGPU gets its outstanding performance over many years of use on the top 
 ten HPC systems in the world. We show that many of the challenges particle
 -in-cell codes face towards Exascale performance are shared with HPC and h
 igh data rate applications in many other domains. By carefully disecting t
 hese common features from the domain of particle-in-cell codes we are now 
 able to reuse our solutions and make them available to scientists across t
 he globe. Now, there is an open, Exascale-ready ecosystem you can use.\n\n
 Domain: Computer Science and Applied Mathematics, Engineering, Physics
END:VEVENT
END:VCALENDAR