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:20220812T074335Z
LOCATION:Nairobi Room
DTSTART;TZID=Europe/Stockholm:20220629T123000
DTEND;TZID=Europe/Stockholm:20220629T130000
UID:submissions.pasc-conference.org_PASC22_sess161_msa250@linklings.com
SUMMARY:Resolving Complex Three-Dimensional Stress Distribution in Earth’s
  Lithosphere using a Julia Multi-GPU Framework
DESCRIPTION:Minisymposium\n\nResolving Complex Three-Dimensional Stress Di
 stribution in Earth’s Lithosphere using a Julia Multi-GPU Framework\n\nMac
 herel, Podladchikov, Räss, Schmalholz\n\nPower-law viscous flow describes 
 the first-order features of long-term lithosphere deformation. Due to the 
 ellipticity of the Earth, the lithosphere is mechanically analogous to a s
 hell, characterized by a double curvature, resulting in fundamentally diff
 erent mechanical characteristics compared to a straight plate. The systema
 tic quantification of the magnitude and the spatiotemporal stress distribu
 tion inside a deforming lithospheric shell is thus of major importance. Th
 e complex shape and heterogeneity of the lithosphere requires a three-dime
 nsional (3D) and high-resolution, thus high-performance computing approach
 .<br />Here, we present numerical simulations of mechanically heterogeneou
 s, incompressible 3D power-law viscous flow under gravity in cartesian, cy
 lindrical and spherical coordinate-systems. We employ the pseudo-transient
  finite-difference (PTFD) method, which enables efficient simulations of h
 igh-resolution 3D deformation processes by implementing an iterative and i
 mplicit solution strategy of the governing equations. The main challenges 
 for the PTFD method are to minimize the iteration count and to ensure an e
 ffective memory throughput as close as possible to hardware’s peak memory 
 throughput. We discuss our PTFD solver implementation using the Julia lang
 uage and the ParallelStencil.jl package, which enables optimal parallel ex
 ecution on GPUs and CPUs, and the ImplicitGlobalGrid.jl package which show
 s ideal scalability up to thousands of GPUs.\n\nDomain: Climate, Weather a
 nd Earth Sciences, Computer Science and Applied Mathematics, Physics
END:VEVENT
END:VCALENDAR