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:20260611T145139Z
LOCATION:Bldg. 6 - Room 002
DTSTART;TZID=Europe/Stockholm:20260701T120000
DTEND;TZID=Europe/Stockholm:20260701T123000
UID:submissions.pasc-conference.org_PASC26_sess172_pap121@linklings.com
SUMMARY:A Comparison of Massively Parallel Performance Portable Particle-i
 n-Cell Schemes for Electrostatic Kinetic Plasma Simulations
DESCRIPTION:Sonali Mayani (Paul Scherrer Institute, ETH Zurich); Paul Fisc
 hill (ETH Zurich); Sriramkrishnan Muralikrishnan (Forschungszentrum Jülich
 ); and Andreas Adelmann (Paul Scherrer Institute, ETH Zurich)\n\nWe compar
 e different Poisson solvers within the context of an electrostatic Vlasov-
 Poisson system. These schemes are implemented as part of the IPPL (Indepen
 dent Parallel Particle Layer) library, which provides performance portable
  and dimension independent building blocks for scientific simulations requ
 iring particle-mesh methods, with Eulerian (mesh-based) and Lagrangian (pa
 rticle-based) approaches. The simulation used to compare the performance a
 nd portability of the schemes is Landau damping, part of a set of mini-app
 lications implemented to benchmark and showcase the capabilities of the IP
 PL library. We use grid-sizes of $512^3$ and $1024^3$ with 8 particles per
  cell, running with different algorithms in the solve phase of the Particl
 e-in-Cell (PIC) loop: a fast Fourier transform (FFT) pseudo-spectral solve
 r, a matrix-free finite difference Conjugate Gradient solver, and a matrix
 -free Finite Element (FEM) solver. We also compare these PIC schemes to th
 e novel Particle-in-Fourier (PIF) scheme, which performs interpolations us
 ing non-uniform FFTs thereby avoiding a grid in the real space. We obtain 
 results on different computing architectures, such as AMD GPUs (LUMI at CS
 C), and Nvidia GPUs (Alps at CSCS and JUWELS Booster at Jülich Supercomput
 ing Center), showcasing portability. In terms of absolute time the FFT sol
 ver is advantageous, but is limited in its applicability. All other field 
 solvers in the PIC scheme are an order-of-magnitude slower, but scale simi
 larly to the FFT case in the electrostatic PIC context. The PIF scheme ser
 ves as a high fidelity alternative to standard PIC, while costlier than th
 e FFT-based PIC scheme, it shows excellent scalability on all the architec
 tures.\n\nDomain: Physics\n\nSession Chair: Andreas Adelmann (Paul Scherre
 r Institute, ETH Zurich)\n\n
END:VEVENT
END:VCALENDAR