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:20250822T115805Z
LOCATION:Campussaal - Plenary Room
DTSTART;TZID=Europe/Stockholm:20250617T103000
DTEND;TZID=Europe/Stockholm:20250617T110000
UID:submissions.pasc-conference.org_PASC25_sess150@linklings.com
SUMMARY:Flash Poster Session - Part II
DESCRIPTION:P37 - pyGinkgo: Python Bindings for Ginkgo\n\nOver the past de
 cade, machine learning has achieved significant advancements, with applica
 tions spanning diverse fields such as physics, medicine, economics or ener
 gy. A pressing challenge in contemporary machine learning is optimizing mo
 dels for time and energy efficiency. One effective approach to...\n\n\nKes
 hvi Tuteja and Gregor Olenik (Karlsruhe Institute of Technology); Roman Mi
 shchuk and Nicolas Venkovic (Technical University of Munich); Markus Götz 
 and Achim Streit (Karlsruhe Institute of Technology); Hartwig Anzt (Techni
 cal University of Munich, University of Tennessee); and Charlotte Debus (K
 arlsruhe Institute of Technology)\n---------------------\nP39 - Simulation
 s of Giant Impacts with Material Strength in pkdgrav3\n\nGiant impacts for
 m the last stage of planet formation and play a key role in determining ma
 ny aspects like the final structure of planetary systems and the masses an
 d compositions of its constituents. A common choice for numerically solvin
 g the equations of motion is the Smoothed Particle Hydrodynam...\n\n\nThom
 as Meier (University of Zurich); Christian Reinhardt (University of Zurich
 , University of Bern); and Douglas Potter and Joachim Stadel (University o
 f Zurich)\n---------------------\nP43 - Towards Exascale Particle-Mesh Met
 hods: A Massively Parallel Performance Portable C++ Particle-in-Cell Frame
 work\n\nWe showcase the Independent Parallel Particle Layer (IPPL), a perf
 ormance portable C++ library for particle-in-cell methods. IPPL makes use 
 of Kokkos (a performance portability abstraction layer), HeFFTe (a library
  for large scale FFTs), and MPI (Message Passing Interface) to deliver a p
 ortable, mas...\n\n\nSonali Mayani (Paul Scherrer Institute, ETH Zurich); 
 Matthias Frey (University of St Andrews); Sriramkrishnan Muralikrishnan (F
 orschungszentrum Jülich); and Ryan Ammann and Andreas Adelmann (Paul Scher
 rer Institute, ETH Zurich)\n---------------------\nP38 - Scalable Genomic 
 Context Analysis with GCsnap2 on HPC Clusters\n\nGCsnap2 Cluster is a scal
 able, Python-based high performance solution for genomic context analysis,
  co-developed by computer and life scientists to overcome the scalability 
 limitations of its predecessor, GCsnap1 Desktop. Leveraging distributed co
 mputing with mpi4py.futures, GCsnap2 Cluster achieved...\n\n\nReto Krummen
 acher and Osman Seckin Simsek (University of Basel); Michèle Leemann, Leil
 a T. Alexander, and Torsten Schwede (University of Basel, Swiss Institute 
 of Bioinformatics); Florina M. Ciorba (University of Basel); and Joana Per
 eira (University of Basel, Swiss Institute of Bioinformatics)\n-----------
 ----------\nP35 - Performance Portability Across Different Mathematical Mo
 dels, Hardware, and Simulation Scenarios in Molecular Dynamics\n\nDue to t
 he importance of Molecular Dynamics simulations within fields such as ther
 modynamics, numerous methods have been developed to speedup the force calc
 ulations, which typically dominate the runtime. None of these methods are,
  however, optimal for every molecular model, on every hardware, and fo...\
 n\n\nSamuel James Newcome, Fabio Alexander Gratl, Manish Kumar Mishra, Mar
 kus Mühlhäußer, Jonas Schumacher, and Hans-Joachim Bungartz (Technical Uni
 versity of Munich)\n---------------------\nP29 - itwinai: Enabling Scalabl
 e AI Workflows on HPC for Digital Twins in Science\n\nThe interTwin projec
 t is advancing the integration of Digital Twins across scientific domains,
  focusing on physics and climate research. A key component of this project
  is itwinai, a Python library designed to streamline scalable AI workflows
  on High-Performance Computing (HPC) systems. With its uni...\n\n\nMatteo 
 Bunino, Anna Elisa Lappe, and Jarl Sondre Sæther (CERN); Rakesh Sarma (FZ 
 Jülich); Maria Girone (CERN); and Andreas Lintermann (Forschungszentrum Jü
 lich)\n---------------------\nP33 - Optimizing Data Offload in the IFS Usi
 ng GPU-Aware Data Structures and Source-To-Source Translation\n\nThe adapt
 ation of the ECMWF’s medium-range forecasting model, the Integrated Foreca
 sting System (IFS), to heterogeneous computing architectures is an ongoing
  effort. The IFS consists of millions of lines of Fortran code that is hig
 hly optimized for modern CPUs. This poses significant challenge...\n\n\nJo
 han Ericsson, Ahmad Nawab, and Balthasar Reuter (ECMWF); Philippe Marguina
 ud and Judicaël Grasset (Meteo-France); and Michael Lange (ECMWF)\n-------
 --------------\nACMP01 - Designing Biomimetic Materials for Carbon Capture
 : Leveraging High-Performance Computing for Large-Scale Molecular Dynamics
  Simulations to Advance Sustainable Solutions\n\nSustainable carbon captur
 e and greenhouse gas mitigation demand innovative strategies that harness 
 biomolecular functions and integrate them into existing technologies. Enzy
 me-based systems offer a promising solution for sustainable CO₂ capture, y
 et their industrial adoption is limited by limi...\n\n\nMerve Fedai and Ya
 roslava Yingling (North Carolina State University)\n---------------------\
 nACMP02 - Development of a Predictive Model for the Prognosis of Patients 
 with Breast Cancer\n\nBreast cancer is one of the most prevalent malignanc
 ies among women, accounting for about a quarter of all new diagnoses world
 wide. Treatment and prognosis vary according to histological subtype and s
 tage at diagnosis. Because of heterogeneity in treatment responses, biomar
 kers that predict clinical...\n\n\nPatricia Honorato Moreira (Inteli - Ins
 titute of Technology and Leadership)\n---------------------\nP28 - Interac
 tive Visualization of High-Energy Physics Events via Nvidia Omniverse\n\nS
 imulations play a crucial role in high energy, nuclear, and accelerator ph
 ysics, aiding in both data analysis and hardware development. Over the yea
 rs, several advanced programs have been created to generate detailed and p
 recise simulated events, providing insights into complex physical processe
 s. ...\n\n\nFelice Nenna (INFN Bari, University of Padova); Marcello Maggi
  (INFN Bari); Matteo Bunino (CERN); Stewart Boogert (University of Manches
 ter); and Siobhan Alden (Royal Holloway, University of London)\n----------
 -----------\nP36 - Pyccel: Automating Translation of Python Prototypes to 
 C/Fortran Production Codes\n\nPython is a widely popular programming langu
 age, valued for its simplicity, ease of learning, and vast ecosystem of pa
 ckages, making it ideal for scientific applications. However, its executio
 n speed is a major limitation compared to low-level languages. We present 
 Pyccel, an intuitive transpiler th...\n\n\nEmily Bourne (EPFL), Mohamed Ja
 lal Maaouni (UM6P), and Yaman Güçlü (Max Planck Institute for Plasma Physi
 cs)\n---------------------\nP30 - The MENTOR Interpretation Agent: From Ne
 twork Embeddings to Mechanistic Narratives via Retrieval-Augmented LLMs\n\
 nDespite an increasing number of complex omics data sets, extracting compr
 ehensive mechanistic insights from these data remains challenging. To addr
 ess this, we developed a human-in-the-loop LLM-based agentic retrieval-aug
 mented generation (RAG) pipeline, the MENTOR Interpretation Agent (MENTOR-
 IA), ...\n\n\nAnna H.C. Vlot (Oak Ridge National Laboratory); Matthew Lane
  (Oak Ridge National Laboratory; Bredesen Center for Interdisciplinary Gra
 duate Research and Education, University of Tennessee-Knoxville); Kyle A. 
 Sullivan (Oak Ridge National Laboratory); Peter Kruse (Oak Ridge National 
 Laboratory; Bredesen Center for Interdisciplinary Graduate Research and Ed
 ucation, University of Tennessee-Knoxville); John Dandy and Selin Kaplanog
 lu (Oak Ridge National Laboratory); Alice Townsend and Jean Merlet (Oak Ri
 dge National Laboratory; Bredesen Center for Interdisciplinary Graduate Re
 search and Education, University of Tennessee-Knoxville); and Daniel A. Ja
 cobson (Oak Ridge National Laboratory)\n---------------------\nP41 - SYCL 
 and Block-Structured Grids: Performance Impact on Simulations of Complex C
 ostal Ocean Domains\n\nDeveloping the next generation of climate modelling
  tools to increase throughput and ensure performance portability is crucia
 l. The choice of an underlying grid for ocean modelling, an important clim
 ate compartment, is difficult. The almost fractal-like boundaries of ocean
  domains and quickly changi...\n\n\nJonathan Schmalfuß (University of Bayr
 euth), Daniel Zint (New York University), Sara Faghih-Naini (ECMWF), Julia
 n Stahl (Friedrich-Alexander-Universität Erlangen-Nürnberg), Markus Büttne
 r (University of Bayreuth), Roberto Grosso (Friedrich-Alexander-Universitä
 t Erlangen-Nürnberg), and Vadym Aizinger (University of Bayreuth)\n-------
 --------------\nP40 - Spectral Methods for the Clustering of Cyclic and Ac
 yclic Graphs\n\nTraditional spectral clustering methods are designed for u
 ndirected graphs and fail to capture the directionality of the edges and o
 f the connections between the clusters. The aim of our work is centered ar
 ound developing novel spectral methods for the spectral clustering of dire
 cted graphs with blo...\n\n\nJacopo Palumbo (Università della Svizzera ita
 liana, Politecnico di Milano); Dimosthenis Pasadakis (Università della Svi
 zzera italiana); Albert-Jan Yzelman (Huawei); and Olaf Schenk (Università 
 della Svizzera italiana)\n---------------------\nACMP05 - A Performance Po
 rtable Matrix-Free Finite Element Framework for Particle-Mesh Methods\n\nC
 omputing architectures are becoming increasingly complex and potent, as we
  reach new computing capacities. Currently the first three machines in the
  TOP500 list are exascale systems. To be able to take full advantage of th
 ese machines, and even run on such heterogeneous architectures, it has bec
 ome...\n\n\nSonali Mayani (Paul Scherrer Institute, ETH Zurich)\n---------
 ------------\nACMP03 - Distributed Computing for Spatio-Temporal Bayesian 
 Modeling Using the INLA Method\n\nBayesian inference on large-scale spatio
 -temporal models is limited by its computational feasibility, a trend that
  is further exacerbated by the continuous increase in data availability an
 d model refinements. To address this issue, we present a double-layer dist
 ributedmemory parallelization strategy...\n\n\nVincent Maillou and Alexand
 ros Nikolaos Ziogas (ETH Zurich); Olaf Schenk (Università della Svizzera i
 taliana); Mathieu Luisier (ETH Zurich); Håvard Rue (King Abdullah Universi
 ty of Science and Technology); and Lisa Gaedke-merzhaeuser (King Abdullah 
 University of Science and Technology, Università della Svizzera italiana)\
 n---------------------\nP32 - Multi-Omic Single Cell Network Perturbation 
 for Phenotypic Prediction\n\nDrug repurposing offers a cost-effective stra
 tegy to identify new applications for existing medications, leveraging est
 ablished safety profiles to accelerate therapeutic development. Advances i
 n computational biology and large-scale multi-omics data enable systematic
  identification of novel therapeu...\n\n\nMatthew Lane (Oak Ridge National
  Laboratory, University of Tennessee); Erica Prates (Oak Ridge National La
 boratory); Alice Townsend and Jean Merlet (Oak Ridge National Laboratory, 
 University of Tennessee); Christiane Alvarez and Alana Wells (Oak Ridge Na
 tional Laboratory); and Daniel Jacobson (Oak Ridge National Laboratory, Un
 iversity of Tennessee)\n---------------------\nP34 - Optimizing the ECsim 
 Plasma Code for Exascale Architectures: GPU Acceleration, Portability, and
  Scalability\n\nThis work presents the adaptation of the plasma code ECsim
  for future exascale architectures. The code has three main blocks called 
 particle movers, moment gathering and field solver. The first two blocks a
 re the most computationally challenging, thus we focused on optimizing the
 m for GPU accelerati...\n\n\nNitin Shukla (CINECA), Elisabetta Boella (E4 
 Computer Engineering), Filippo Spiga (NVIDIA Inc.), Michael Redenti (CINEC
 A), Mozhgan Kabiri Chimeh (NVIDIA Inc.), and Maria Elena Innocenti (Ruhr U
 niversity Bochum)\n---------------------\nACMP06 - Understanding HMM Perfo
 rmance for Enhanced HPC Portability\n\nHeterogeneous Memory Management (HM
 M) simplifies programming for heterogeneous systems, making High-Performan
 ce Computing (HPC) devices more accessible to domain scientists; however, 
 it suffers from slow performance compared to other memory management appro
 aches. HMM is an infrastructure provided by...\n\n\nNicholas Cassarino (Un
 iversity of North Carolina at Charlotte)\n---------------------\nP27 - Int
 egrating the ICON4Py Python-Based Dynamical Core into ICON\n\nThe integrat
 ion of Python-based high-performance computing into legacy Fortran climate
  models offers new opportunities for flexibility and efficiency. This post
 er presents the integration, in the Fortran ICON implementation of the dyn
 amical core implemented in Python, as part of ICON4py, a still in-...\n\n\
 nMauro Bianco (ETH Zurich / CSCS), Magdalena Luz (ETH Zurich), Christoph M
 uller and Daniel Hupp (MeteoSwiss), Anurag Dipankar (ETH Zurich), Edoardo 
 Paone (ETH Zurich / CSCS), Xavier Lapillonne (MeteoSwiss), Nicoletta Farab
 ullini (ETH Zurich), Enrique Gonzales Pareder and Hannes Vogt (ETH Zurich 
 / CSCS), Ong Chia Rui (ETH Zurich), Till Ehrengruber (ETH Zurich / CSCS), 
 Yilu Chen (ETH Zurich), and Philip Muller and Christos Kotsalos (ETH Zuric
 h / CSCS)\n---------------------\nP31 - Mixed Precision Customized for Dis
 continuous Galerkin Methods\n\nWe present an approach to enhance storage e
 fficiency and reduce memory bandwidth utilization in modal Discontinuous G
 alerkin (DG) methods by introducing a customized mixed-precision\nrepresen
 tation for the solution vector. Our approach leverages variations in float
 ing-point accuracy requirements amon...\n\n\nShivam Sundriyal, Markus Bütt
 ner, and Vadym Aizinger (University of Bayreuth)\n\nSession Chair: David M
 oxey (King's College London)
END:VEVENT
END:VCALENDAR