{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,9,19]],"date-time":"2025-09-19T11:31:31Z","timestamp":1758281491028,"version":"3.44.0"},"reference-count":24,"publisher":"Association for Computing Machinery (ACM)","issue":"3","content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Math. Softw."],"published-print":{"date-parts":[[2025,9,30]]},"abstract":"<jats:p>Vico et al. suggest a fast algorithm for computing volume potentials, beneficial to fields with problems requiring the solution of the free-space Poisson\u2019s equation, such as beam and plasma physics. Currently, the standard is the algorithm of Hockney and Eastwood, with second order in convergence at best. The algorithm proposed by Vico et\u00a0al. converges spectrally for sufficiently smooth functions, i.e., faster than any fixed order in the number of grid points. We implement a performance portable version of the traditional Hockney-Eastwood and the novel Vico-Greengard Poisson solver as part of the Independent Parallel Particle Layer (IPPL) library. For sufficiently smooth source functions, the Vico-Greengard algorithm achieves higher accuracy than the Hockney-Eastwood method with the same grid size, reducing the computational demands of high-resolution simulations since one could use coarser grids to achieve them. Additionally, we propose an improvement to the Vico-Greengard method which further reduces its memory footprint. This is important for GPUs, which have limited memory, and should be taken into account when selecting numerical algorithms for performance portable codes. Finally, we showcase performance through GPU and CPU scaling studies on the Perlmutter (NERSC) supercomputer, with efficiencies staying above 50% in the strong scaling case. To showcase portability, we also run the scaling studies on the Alps supercomputer at CSCS, Switzerland and the GPU partition of the Lumi supercomputer at CSC, Finland.<\/jats:p>","DOI":"10.1145\/3748815","type":"journal-article","created":{"date-parts":[[2025,7,16]],"date-time":"2025-07-16T14:21:42Z","timestamp":1752675702000},"page":"1-23","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["A Massively Parallel Performance Portable Free-Space Spectral Poisson Solver"],"prefix":"10.1145","volume":"51","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-3294-4650","authenticated-orcid":false,"given":"Sonali","family":"Mayani","sequence":"first","affiliation":[{"name":"Paul Scherrer Institute, Villigen, Switzerland and ETH Zurich, Zurich, Switzerland"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0004-5772-6109","authenticated-orcid":false,"given":"Veronica","family":"Montanaro","sequence":"additional","affiliation":[{"name":"ETH Zurich, Zurich, Switzerland"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3326-4785","authenticated-orcid":false,"given":"Antoine","family":"Cerfon","sequence":"additional","affiliation":[{"name":"Type One Energy Group - Canada Inc., Vancouver, British Columbia, Canada"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7842-0051","authenticated-orcid":false,"given":"Matthias","family":"Frey","sequence":"additional","affiliation":[{"name":"Mathematical Institute, University of St Andrews, St Andrews, UK"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5494-1880","authenticated-orcid":false,"given":"Sriramkrishnan","family":"Muralikrishnan","sequence":"additional","affiliation":[{"name":"J\u00fclich Supercomputing Centre, Forschungszentrum J\u00fclich GmbH, J\u00fclich, Germany"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7230-7007","authenticated-orcid":false,"given":"Andreas","family":"Adelmann","sequence":"additional","affiliation":[{"name":"Paul Scherrer Institute, Villigen, Switzerland"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"320","published-online":{"date-parts":[[2025,9,18]]},"reference":[{"unstructured":"Andreas Adelmann Pedro Calvo Matthias Frey Achim Gsell Uldis Locans Christof Metzger-Kraus Nicole Neveu Chris Rogers Steve Russell Suzanne Sheehy et al. 2019. OPAL a versatile tool for charged particle accelerator simulations. arXiv:1905.06654. Retrieved July 15 2021 from http:\/\/arxiv.org\/abs\/1905.06654","key":"e_1_3_3_2_1"},{"doi-asserted-by":"publisher","key":"e_1_3_3_3_1","DOI":"10.1007\/978-3-030-50371-0_19"},{"key":"e_1_3_3_4_1","volume-title":"Analysis of the Communication and Computation Cost of FFT Libraries towards Exascale","author":"Ayala Alan","year":"2022","unstructured":"Alan Ayala, Stanimire Tomov, Piotr Luszczek, Sebastien Cayrols, Gerald Ragghianti, and Jack Dongarra. 2022. Analysis of the Communication and Computation Cost of FFT Libraries towards Exascale. Technical Report. Innovative Computing Laboratory."},{"doi-asserted-by":"publisher","key":"e_1_3_3_5_1","DOI":"10.1016\/j.cpc.2011.05.014"},{"doi-asserted-by":"publisher","key":"e_1_3_3_6_1","DOI":"10.1016\/j.jpdc.2014.07.003"},{"unstructured":"Matthias Frey Alessandro Vinciguerra Sriramkrishnan Muralikrishnan Sonali Mayani Veronica Montanaro and Andreas Adelmann. 2023. IPPL-framework\/ippl: IPPL-3.1.0. Retrieved March 4 2024 from https:\/\/zenodo.org\/records\/8389193","key":"e_1_3_3_7_1"},{"doi-asserted-by":"publisher","key":"e_1_3_3_8_1","DOI":"10.5555\/62815"},{"doi-asserted-by":"publisher","key":"e_1_3_3_9_1","DOI":"10.1016\/0021-9991(77)90013-4"},{"unstructured":"kokkos\/kokkos-tools. 2025. Kokkos C++ Performance Portability Programming Ecosystem: Profiling and Debugging Tools. Retrieved January 31 2025 from https:\/\/github.com\/kokkos\/kokkos-tools\/tree\/develop","key":"e_1_3_3_10_1"},{"doi-asserted-by":"publisher","key":"e_1_3_3_11_1","DOI":"10.1137\/23M1550803"},{"unstructured":"LLNL\/mpiP. 2025. LLNL\/mpiP: A Light-Weight MPI Profiler. Retrieved 20 May from https:\/\/github.com\/LLNL\/mpiP","key":"e_1_3_3_12_1"},{"unstructured":"Veronica Montanaro. 2023. Improvements to the State-of-the-Art Open Boundary FFT Poisson Solver. Retrieved January 28 2025 from https:\/\/amas.web.psi.ch\/people\/aadelmann\/ETH-Accel-Lecture-1\/projectscompleted\/cse\/Montanaro_report_final.pdf","key":"e_1_3_3_13_1"},{"doi-asserted-by":"publisher","key":"e_1_3_3_14_1","DOI":"10.1137\/1.9781611977967.3"},{"doi-asserted-by":"publisher","key":"e_1_3_3_15_1","DOI":"10.1103\/PhysRevAccelBeams.22.054602"},{"doi-asserted-by":"publisher","key":"e_1_3_3_16_1","DOI":"10.1103\/PhysRevSTAB.9.044204"},{"unstructured":"Kenneth J. Raffenetti. 2024. MPICH 4.1.2 Released|MPICH. Retrieved January 9 2024 from https:\/\/www.mpich.org\/2023\/06\/08\/mpich-4-1-2-released\/","key":"e_1_3_3_17_1"},{"unstructured":"Johannes Toph\u00f8j Rasmussen. 2011. Particle Methods in Bluff Body Aerodynamics. DCAMM Report. DTU Mechanical Engineering.","key":"e_1_3_3_18_1"},{"unstructured":"Robert D. Ryne. 2011. On FFT-based convolutions and correlations with application to solving Poisson\u2019s equation in an open rectangular pipe. arXiv:1111.4971. Retrieved April 22 2021 from http:\/\/arxiv.org\/abs\/1111.4971","key":"e_1_3_3_19_1"},{"doi-asserted-by":"publisher","key":"e_1_3_3_20_1","DOI":"10.1109\/PDP.2011.42"},{"unstructured":"Miroslav Stoyanov. 2024. Private Communication. We talked about the heFFTe library\u2019s scaling behaviour on the GH200 architecture and it was suggested that heFFTe is not optimizing for this hardware yet.","key":"e_1_3_3_21_1"},{"doi-asserted-by":"publisher","key":"e_1_3_3_22_1","DOI":"10.1088\/0741-3335\/47\/5A\/017"},{"doi-asserted-by":"publisher","key":"e_1_3_3_23_1","DOI":"10.1016\/j.jcp.2016.07.028"},{"unstructured":"Eric W. Weisstein. 2021. Convolution Theorem. Wolfram Research Inc. Retrieved August 12 2021 from https:\/\/mathworld.wolfram.com\/ConvolutionTheorem.html","key":"e_1_3_3_24_1"},{"unstructured":"Junyi Zou Eugenia Kim and Antoine J. Cerfon. 2021. FFT-based free space Poisson solvers: Why Vico-Greengard-Ferrando should replace Hockney-Eastwood. arXiv:2103.08531. Retrieved Apr. 22 2021 from http:\/\/arxiv.org\/abs\/2103.08531","key":"e_1_3_3_25_1"}],"container-title":["ACM Transactions on Mathematical Software"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3748815","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,9,19]],"date-time":"2025-09-19T02:49:39Z","timestamp":1758250179000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3748815"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,9,18]]},"references-count":24,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2025,9,30]]}},"alternative-id":["10.1145\/3748815"],"URL":"https:\/\/doi.org\/10.1145\/3748815","relation":{},"ISSN":["0098-3500","1557-7295"],"issn-type":[{"type":"print","value":"0098-3500"},{"type":"electronic","value":"1557-7295"}],"subject":[],"published":{"date-parts":[[2025,9,18]]},"assertion":[{"value":"2024-06-26","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-07-04","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-09-18","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}