{"paper":{"title":"GPU acceleration of plane-wave density functional theory calculations in Abinit","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Abinit achieves GPU speedups for plane-wave DFT by revising the Kohn-Sham iterative diagonalizer.","cross_cats":[],"primary_cat":"cond-mat.mtrl-sci","authors_text":"Ioanna-Maria Lygatsika, Lucas Baguet, Marc Sarraute, Marc Torrent, Pierre Kestener","submitted_at":"2026-04-13T07:54:17Z","abstract_excerpt":"We report on the GPU port of the Abinit high-performance simulation code for plane-wave DFT calculations. Large-scale electronic structure calculations require computing the electronic wave function by solving the Kohn-Sham equations discretized over a large number of plane waves. Porting such calculations to GPU nodes relies not only on extensive usage of vendor libraries from a development perspective, but also on algorithmic revisions of the iterative diagonalization procedure in the resolution of the Kohn-Sham equations to identify GPU-efficient mathematical operations (linear algebra, FFT"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"The Abinit implementation on multi-GPU architectures provides detailed performance results to compare CPU nodes versus heterogeneous CPU-GPU nodes, with particular attention given to the comparison of LOBPCG and Chebyshev polynomial filtering in terms of their GPU efficiency.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the algorithmic revisions to the iterative diagonalization procedure preserve the numerical accuracy and convergence properties of the original CPU implementation while achieving the reported GPU speedups.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Abinit's plane-wave DFT solver has been ported to multi-GPU architectures with performance benchmarks comparing CPU and CPU-GPU nodes for LOBPCG and Chebyshev filtering diagonalization methods.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Abinit achieves GPU speedups for plane-wave DFT by revising the Kohn-Sham iterative diagonalizer.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"95e2dee9dfeaddc696847923e8aa852b5c9d237ac73101859dcd5519a7e0a052"},"source":{"id":"2604.11139","kind":"arxiv","version":2},"verdict":{"id":"69ab9b7d-b087-4610-9fc6-314ff24a5503","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-10T15:00:26.235557Z","strongest_claim":"The Abinit implementation on multi-GPU architectures provides detailed performance results to compare CPU nodes versus heterogeneous CPU-GPU nodes, with particular attention given to the comparison of LOBPCG and Chebyshev polynomial filtering in terms of their GPU efficiency.","one_line_summary":"Abinit's plane-wave DFT solver has been ported to multi-GPU architectures with performance benchmarks comparing CPU and CPU-GPU nodes for LOBPCG and Chebyshev filtering diagonalization methods.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the algorithmic revisions to the iterative diagonalization procedure preserve the numerical accuracy and convergence properties of the original CPU implementation while achieving the reported GPU speedups.","pith_extraction_headline":"Abinit achieves GPU speedups for plane-wave DFT by revising the Kohn-Sham iterative diagonalizer."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.11139/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}