{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:O3HCAAG5UAO4N3D5NHCJ2O577N","short_pith_number":"pith:O3HCAAG5","schema_version":"1.0","canonical_sha256":"76ce2000dda01dc6ec7d69c49d3bbffb6ee9fff292d99a7e5678dd80ff2bbe33","source":{"kind":"arxiv","id":"2504.13821","version":1},"attestation_state":"computed","paper":{"title":"Toward Portable GPU Performance: Julia Recursive Implementation of TRMM and TRSM","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC"],"primary_cat":"cs.MS","authors_text":"Alan Edelman, Evelyne Ringoot, James Schloss, Maxwell Onyango, Rabab Alomairy, Vicki Carrica","submitted_at":"2025-04-18T17:51:34Z","abstract_excerpt":"This paper presents a performant and portable recursive implementation of triangular matrix-matrix multiplication (TRMM) and triangular solve (TRSM) in Julia for GPUs, two kernels that underlie many linear-algebra algorithms. We restructure TRMM and TRSM so that most work is executed as general matrix-matrix multiplication (GEMM), improving use of the GPU memory hierarchy and reducing latency. Exploiting Julia's multiple dispatch and metaprogramming together with the GPUArrays and KernelAbstractions frameworks, we expose a single hardware-agnostic API that runs on NVIDIA, AMD, and Apple Silico"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2504.13821","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.MS","submitted_at":"2025-04-18T17:51:34Z","cross_cats_sorted":["cs.DC"],"title_canon_sha256":"c42ad5598b0525d8ecf7b79305049155f4d4a8070ba1cf8ceb77fda1609a24f9","abstract_canon_sha256":"6466e96da4dc1f81ebcc906ee34b63f474c9b4c361c2586829ef833713f83538"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:51:02.588017Z","signature_b64":"f5dcxQovQq8ZfIAUMMKNn62WPaA9SrsQdCjOzQpexnawD1MBq+SHp8Li8fEFgS9TZOoldzxqROftuDXhz8k0Cg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"76ce2000dda01dc6ec7d69c49d3bbffb6ee9fff292d99a7e5678dd80ff2bbe33","last_reissued_at":"2026-07-05T10:51:02.587514Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:51:02.587514Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Toward Portable GPU Performance: Julia Recursive Implementation of TRMM and TRSM","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC"],"primary_cat":"cs.MS","authors_text":"Alan Edelman, Evelyne Ringoot, James Schloss, Maxwell Onyango, Rabab Alomairy, Vicki Carrica","submitted_at":"2025-04-18T17:51:34Z","abstract_excerpt":"This paper presents a performant and portable recursive implementation of triangular matrix-matrix multiplication (TRMM) and triangular solve (TRSM) in Julia for GPUs, two kernels that underlie many linear-algebra algorithms. We restructure TRMM and TRSM so that most work is executed as general matrix-matrix multiplication (GEMM), improving use of the GPU memory hierarchy and reducing latency. Exploiting Julia's multiple dispatch and metaprogramming together with the GPUArrays and KernelAbstractions frameworks, we expose a single hardware-agnostic API that runs on NVIDIA, AMD, and Apple Silico"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.13821","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2504.13821/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2504.13821","created_at":"2026-07-05T10:51:02.587571+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.13821v1","created_at":"2026-07-05T10:51:02.587571+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.13821","created_at":"2026-07-05T10:51:02.587571+00:00"},{"alias_kind":"pith_short_12","alias_value":"O3HCAAG5UAO4","created_at":"2026-07-05T10:51:02.587571+00:00"},{"alias_kind":"pith_short_16","alias_value":"O3HCAAG5UAO4N3D5","created_at":"2026-07-05T10:51:02.587571+00:00"},{"alias_kind":"pith_short_8","alias_value":"O3HCAAG5","created_at":"2026-07-05T10:51:02.587571+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.18411","citing_title":"Numerically Stable Cholesky-QR on GPU via Mixed-Precision Randomized Preconditioning","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2606.18411","citing_title":"Numerically Stable Cholesky-QR on GPU via Mixed-Precision Randomized Preconditioning","ref_index":31,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/O3HCAAG5UAO4N3D5NHCJ2O577N","json":"https://pith.science/pith/O3HCAAG5UAO4N3D5NHCJ2O577N.json","graph_json":"https://pith.science/api/pith-number/O3HCAAG5UAO4N3D5NHCJ2O577N/graph.json","events_json":"https://pith.science/api/pith-number/O3HCAAG5UAO4N3D5NHCJ2O577N/events.json","paper":"https://pith.science/paper/O3HCAAG5"},"agent_actions":{"view_html":"https://pith.science/pith/O3HCAAG5UAO4N3D5NHCJ2O577N","download_json":"https://pith.science/pith/O3HCAAG5UAO4N3D5NHCJ2O577N.json","view_paper":"https://pith.science/paper/O3HCAAG5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.13821&json=true","fetch_graph":"https://pith.science/api/pith-number/O3HCAAG5UAO4N3D5NHCJ2O577N/graph.json","fetch_events":"https://pith.science/api/pith-number/O3HCAAG5UAO4N3D5NHCJ2O577N/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/O3HCAAG5UAO4N3D5NHCJ2O577N/action/timestamp_anchor","attest_storage":"https://pith.science/pith/O3HCAAG5UAO4N3D5NHCJ2O577N/action/storage_attestation","attest_author":"https://pith.science/pith/O3HCAAG5UAO4N3D5NHCJ2O577N/action/author_attestation","sign_citation":"https://pith.science/pith/O3HCAAG5UAO4N3D5NHCJ2O577N/action/citation_signature","submit_replication":"https://pith.science/pith/O3HCAAG5UAO4N3D5NHCJ2O577N/action/replication_record"}},"created_at":"2026-07-05T10:51:02.587571+00:00","updated_at":"2026-07-05T10:51:02.587571+00:00"}