BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Chicago
X-LIC-LOCATION:America/Chicago
BEGIN:DAYLIGHT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
TZNAME:CDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
TZNAME:CST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260202T201803Z
LOCATION:263-264
DTSTART;TZID=America/Chicago:20251120T143600
DTEND;TZID=America/Chicago:20251120T145800
UID:submissions.supercomputing.org_SC25_sess177_pap956@linklings.com
SUMMARY:Kilometer-Scale AI-Powered and Performance-Portable Earth System M
 odel (AP3ESM) to Achieve Year-Scale Simulation Speed on Heterogeneous Supe
 rcomputers
DESCRIPTION:Kai Xu, Maoxue Yu, and Yuhu Chen (Laoshan Laboratory); Jie Gao
  (PIESAT Information Technology Co., Ltd.; State Key Laboratory of Numeric
 al Modeling for Atmospheric Sciences and Geophysical Fluid Dynamics, Insti
 tute of Atmospheric Physics, Chinese Academy of Sciences); Shuang Wang and
  Jiaying Song (Gosci Technology Group); Xiaohui Duan (Shandong University)
 ; Junlin Wei (Computer Network Information Center, Chinese Academy of Scie
 nces, Beijing, China; Pengcheng Laboratory, Shenzhen, China; University of
  Chinese Academy of Sciences); Jiangfeng Yu (Institute of Atmospheric Phys
 ics, Chinese Academy of Science，Beijing, China; University of Chinese Acad
 emy of Sciences); Hailong Liu (Laoshan Laboratory); Jinrong Jiang (Compute
 r Network Information Center, Chinese Academy of Sciences); Yi Zhang (Chin
 ese Academy of Meteorological Sciences); Pengfei Lin (Institute of Atmosph
 eric Physics, Chinese Academy of Science); Tianyi Wang (School of Informat
 ion Science and Engineering, Shandong Normal University); Pengfei Wang and
  Weipeng Zheng (Institute of Atmospheric Physics, Chinese Academy of Scien
 ce); Jingwei Xie (Laoshan Laboratory); Jiakang Zhang (Institute of Atmosph
 eric Physics, Chinese Academy of Science，Beijing, China; University of Chi
 nese Academy of Sciences); Zilu Liu and Xiaoyu Jin (Laoshan Laboratory); J
 ilin Wei (Institute of Atmospheric Physics, Chinese Academy of Science); Q
 ixin Chang (Shandong University); Qingxia Lin (Gosci Technology Group); Ya
 nzhi Zhou (Institute of Atmospheric Physics, Chinese Academy of Science，Be
 ijing, China; University of Chinese Academy of Sciences); Weiguo Liu (Shan
 dong University); Wei Xue (Tsinghua University, National Supercomputing Ce
 nter in Wuxi); Yiwen Li (School of Ocean Sciences, China University of Geo
 sciences); Haohuan Fu (Tsinghua University, National Supercomputing Center
  in Wuxi); Yue Yu (Pengcheng Laboratory); Xuebin Chi (Computer Network Inf
 ormation Center, Chinese Academy of Sciences); and Lixin Wu (Laoshan Labor
 atory)\n\nKilometer-scale Earth system models (ESMs) necessitate exascale 
 supercomputers to facilitate realistic simulations of weather phenomena an
 d climate variability over a time span ranging from days to decades. We pr
 esent AP3ESM, an ultra-high-resolution, AI-Powered, Performance-Portable E
 SM coupling atmosphere, land surface, ocean, and sea ice components. By le
 veraging the performance portability features of Kokkos and OpenMP, the AP
 3ESM operates efficiently on two heterogeneous systems while incurring min
 imal development overhead. Advanced optimization techniques, such as adapt
 ive parallel algorithms, AI-enhanced physical parameterizations, and mixed
 -precision computations, have been implemented to further boost the comput
 ational efficiency. The standalone atmospheric and oceanic components of A
 P3ESM each reach 1-km resolution, attaining 0.60 and 1.98 simulated-years-
 per-day (SYPD) using 34.1 million cores and 16,085 GPUs, respectively. The
  holistic AP3ESM (3-km atmosphere, 2-km ocean) sustains 1.01 SYPD on 36.6 
 million cores. Notably, the forecast experiment successfully captures Supe
 r Typhoon Doksuri in 2023 and its associated extreme rainfall across China
 .\n\nTag: Applications, Gordon Bell Climate Finalist\n\nRecording: Livestr
 eamed, Recorded\n\nRegistration Category: Technical Program Reg Pass\n\nSe
 ssion Chair: Shelby Lockhart (Advanced Micro Devices, Inc. (AMD))\n\n
END:VEVENT
END:VCALENDAR