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:20260202T201357Z
LOCATION:231
DTSTART;TZID=America/Chicago:20251117T090000
DTEND;TZID=America/Chicago:20251117T173000
UID:submissions.supercomputing.org_SC25_sess207@linklings.com
SUMMARY:2025 International Workshop on Performance, Portability, and Produ
 ctivity in HPC (P3HPC)
DESCRIPTION:The Performance, Portability, and Productivity in HPC workshop
  aims to bring together developers and researchers with an interest in pra
 ctical solutions, technologies, tools, and methodologies that enable the d
 evelopment of performance-portable applications across a diverse set of cu
 rrent and future high‑performance computers. The topic of Performance, Por
 tability, and Productivity focuses on enabling applications and libraries 
 to run across multiple architectures without significant impact on achieve
 d performance and with the goal of maintaining developer productivity. Thi
 s workshop provides a forum for discussions of successes and failures in t
 ackling the compelling problems that lie at the intersection of performanc
 e, portability, and productivity in high-performance computing. This area 
 touches on many aspects of HPC/AI software development and the workshop pr
 ogram is expected to reflect a wide range of experiences and perspectives,
  including those of compiler, language and runtime experts; applications d
 evelopers and performance engineers; and domain scientists. For more infor
 mation see: https://p3hpc.org/workshop/2025/\n\nComparing Cross-Platform P
 erformance via Node-to-Node Scaling Studies\n\nDue to the increasing diver
 sity of high-performance computing architectures, researchers and practiti
 oners are increasingly interested in comparing a code’s performance and sc
 alability across different platforms. However, there is a lack of availabl
 e guidance on how to actually set up and anal...\n\n\nKenneth Weiss (Lawre
 nce Livermore National Laboratory)\n---------------------\nLunch break (on
  your own)\n---------------------\nPerformance portable batched linear alg
 ebra kernels for transport sweeps using Kokkos\n\nThis paper describes the
  development of performance portable\nbatched linear algebra kernels for S
 N-DG neutron transport\nsweeps using Kokkos. We establish a new sweep algo
 rithm\nfor GPUs that relies on batched linear algebra kernels. We\nimpleme
 nt an optimized batched gesv solver for small linear\nsyste...\n\n\nGabrie
 l Suau (CEA Saclay), Thierry Gautier (Inria), Ansar Calloo and Rémi Baron 
 (CEA Saclay), and Romain Le Tellier (CEA Cadarache)\n---------------------
 \nPreserving CUDA Syntax for SYCL Portability: A Thin C++ Abstraction with
 out Kernel Migration\n\nPreparing large-scale scientific applications for 
 diverse GPU architectures requires strategies that balance performance, po
 rtability, and long-term maintainability.  \nWe introduce a unified kernel
  abstraction and evaluate it using CRK-HACC, a production N-body cosmology
  code, enabling single-source...\n\n\nEsteban M. Rangel and Humza Qureshi 
 (Argonne National Laboratory (ANL))\n---------------------\n2025 Internati
 onal Workshop on Performance, Portability, and Productivity in HPC\n\nThe 
 Performance, Portability, and Productivity in HPC workshop aims to bring t
 ogether developers and researchers with an interest in practical solutions
 , technologies, tools, and methodologies that enable the development of pe
 rformance-portable applications across a diverse set of current and future
 ...\n\n\nCJ Newburn (NVIDIA Corporation), Scott Parker (Argonne National L
 aboratory (ANL)), John Pennycook (.), and Kenneth Weiss (Lawrence Livermor
 e National Laboratory (LLNL))\n---------------------\nHow does abstraction
  enable Performance, Portability, and Productivity?\n\nHigher-level abstra
 ctions enable greater implementation freedom, which can be harnessed to yi
 eld greater portability and performance. At the same time, that higher lev
 el of abstraction can increase productivity through fewer lines of code, l
 ess learning, greater rigor, and fewer bugs.\nThis panel wil...\n\n\nCJ Ne
 wburn (NVIDIA Corporation); Jan Ciesko (Sandia National Laboratories); Ste
 phen Jones (NVIDIA Corporation); Rio Yokota (Institute for Science, Tokyo)
 ; and Yanbo Zhao (North Carolina State University)\n---------------------\
 nMixed-Precision Performance Portability of FFT-Based GPU-Accelerated Algo
 rithms for Block-Triangular Toeplitz Matrices\n\nThe hardware diversity in
  leadership-class computing facilities, alongside the immense performance 
 boosts from today's GPUs when computing in lower precision, incentivizes s
 cientific HPC workflows to adopt mixed-precision algorithms and performanc
 e portability models. We present an on-the-fly framew...\n\n\nSreeram Venk
 at (The University of Texas at Austin; Advanced Micro Devices, Inc. (AMD))
 ; Kasia Swirydowicz and Noah Wolfe (Advanced Micro Devices, Inc. (AMD)); a
 nd Omar Ghattas (The University of Texas at Austin)\n---------------------
 \nLAMMPS-KOKKOS: Performance Portable Molecular Dynamics Across Exascale A
 rchitectures\n\nSince its inception in 1995, LAMMPS has grown to be a worl
 d-class molecular dynamics code, with thousands of users, over one million
  lines of code, and multi-scale simulation capabilities. We discuss how LA
 MMPS has adapted to the modern heterogeneous computing landscape by integr
 ating the Kokkos per...\n\n\nAnders Johansson (Sandia National Laboratorie
 s); Evan Weinberg (NVIDIA Corporation); and Christian Trott, Megan McCarth
 y, and Stan Moore (Sandia National Laboratories)\n---------------------\nD
 evelopment of a performance portable distributed FFT interface on top of t
 he Kokkos ecosystem\n\nThis paper presents the development of a performanc
 e portable distributed FFT implementation on top of the Kokkos ecosystem. 
 Thanks to kokkos and kokkos-fft, we largely simplify the implementation de
 tails of distributed FFT with performance portability. We newly develop un
 ique features like batched-...\n\n\nYuuichi Asahi and Trévis Morvany (CEA 
 Saclay, Maison de la Simulation) and Thomas Padioleau and Julien Bigot (CE
 A Saclay)\n---------------------\nEnergy-aware performance portability wit
 h OpenMP dynamic variants\n\nPerformance portability in HPC and embedded s
 ystems is often limited by power and thermal constraints. The OpenMP progr
 amming model offers a compile-time mechanism known as variants, allowing d
 ifferent function specializations. Previous research extended this concept
  to the runtime level, enabling d...\n\n\nAyman Bourramouss el Maach, Adri
 an Munera, and Sara Royuela (Barcelona Supercomputing Center (BSC))\n-----
 ----------------\nRoofline Analysis of Tightly-Coupled CPU-GPU Superchips:
  A Study on MI300A and GH200\n\nThe introduction of tightly-coupled hetero
 geneous architectures, such as AMD's MI300A and  NVIDIA's Grace-Hopper(GH2
 00), address a bottleneck in accelerated computing, namely the CPU-GPU int
 erface.  \nWhereas the GH200 can be seen as a technological leap in CPU-GP
 U connectivity greatly exceeding PCIe...\n\n\nOscar Antepara, Leonid Olike
 r, and Samuel Williams (Lawrence Berkeley National Laboratory (LBNL))\n---
 ------------------\nExtending RAJA Parallel Programming Abstractions with 
 Just-In-Time Optimization\n\nThe prevalence of heterogeneous computing sys
 tems -- comprising both CPUs and GPUs -- has led to the adoption of perfor
 mance portability programming models, such as RAJA. These models allow dev
 elopers to write portable code that compiles ahead-of-time (AOT), unmodifi
 ed for different backends, thus i...\n\n\nJohn Bowen, Konstantinos Parasyr
 is, David Beckingsale, Tal Ben-Nun, Thomas Stitt, and Giorgis Georgakoudis
  (Lawrence Livermore National Laboratory (LLNL))\n---------------------\nA
 fternoon Break - Workshop on Performance, Portability, and Productivity in
  HPC\n---------------------\nBridging Performance Portability and Scalabil
 ity for Plasma Simulations on Heterogeneous Systems\n\nBridging portabilit
 y and scalability is essential for HPC applications. The Vector Particle-I
 n-Cell (VPIC) code, widely used in plasma physics simulations, historicall
 y required extensive platform-specific optimizations to achieve high perfo
 rmance. VPIC 2.0 addresses this challenge by adopting Kokko...\n\n\nNigel 
 P. Tan (Los Alamos National Laboratory (LANL)); Scott V. Luedtke (Los Alam
 os National Laboratory); Michela Taufer (University of Tennessee, Knoxvill
 e); and Brian Albright (Los Alamos National Laboratory (LANL))\n----------
 -----------\nMorning Break - Workshop on Performance, Portability, and Pro
 ductivity in HPC\n\nTag: Performance Evaluation, Scalability, & Portabilit
 y\n\nRecording: Livestreamed, Recorded\n\nRegistration Category: Technical
  Program Reg Pass, Workshop Reg Pass\n\nSession Chairs: CJ Newburn (NVIDIA
  Corporation), Scott J. Parker (Argonne National Laboratory (ANL)), John P
 ennycook (Intel Corporation), and Kenneth Weiss (Lawrence Livermore Nation
 al Laboratory (LLNL))
END:VEVENT
END:VCALENDAR
