{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:FBF3MTBHSPI4HVAQLJ733QP7T6","short_pith_number":"pith:FBF3MTBH","schema_version":"1.0","canonical_sha256":"284bb64c2793d1c3d4105a7fbdc1ff9fb3b45383c06273a4d25fae285b111789","source":{"kind":"arxiv","id":"2507.15121","version":2},"attestation_state":"computed","paper":{"title":"AMPED: Accelerating MTTKRP for Billion-Scale Sparse Tensor Decomposition on Multiple GPUs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.DC","authors_text":"Rajgopal Kannan, Sasindu Wijeratne, Viktor Prasanna","submitted_at":"2025-07-20T21:01:13Z","abstract_excerpt":"Matricized Tensor Times Khatri-Rao Product (MTTKRP) is the computational bottleneck in sparse tensor decomposition. As real-world sparse tensors grow to billions of nonzeros, they increasingly demand higher memory capacity and compute throughput from hardware accelerators. In this work, we present AMPED, a multi-GPU parallel algorithm designed to accelerate MTTKRP on billion-scale sparse tensors. AMPED scales beyond the limits of a single GPU, meeting both the memory and performance requirements of large-scale workloads. We introduce a partitioning strategy combined with a dynamic load balanci"},"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":"2507.15121","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2025-07-20T21:01:13Z","cross_cats_sorted":[],"title_canon_sha256":"dc55faae54573ce7c1103c005849773d2af901bbd2894f37437e6636e4962045","abstract_canon_sha256":"bcc87eabcb1806fac13c2a69b7e87395779f511f1c60c5b7639fefab67cbcb6b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:51:32.376122Z","signature_b64":"gL9/rWGwcKXJJsTqKWW3CxvISNrJBU43EoAE5KbGIYWL5fAVix/1D4oT3i+9IPBQPo4JkmAAyYrphSZZpGlQDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"284bb64c2793d1c3d4105a7fbdc1ff9fb3b45383c06273a4d25fae285b111789","last_reissued_at":"2026-07-05T11:51:32.375606Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:51:32.375606Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"AMPED: Accelerating MTTKRP for Billion-Scale Sparse Tensor Decomposition on Multiple GPUs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.DC","authors_text":"Rajgopal Kannan, Sasindu Wijeratne, Viktor Prasanna","submitted_at":"2025-07-20T21:01:13Z","abstract_excerpt":"Matricized Tensor Times Khatri-Rao Product (MTTKRP) is the computational bottleneck in sparse tensor decomposition. As real-world sparse tensors grow to billions of nonzeros, they increasingly demand higher memory capacity and compute throughput from hardware accelerators. In this work, we present AMPED, a multi-GPU parallel algorithm designed to accelerate MTTKRP on billion-scale sparse tensors. AMPED scales beyond the limits of a single GPU, meeting both the memory and performance requirements of large-scale workloads. We introduce a partitioning strategy combined with a dynamic load balanci"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.15121","kind":"arxiv","version":2},"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/2507.15121/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":"2507.15121","created_at":"2026-07-05T11:51:32.375675+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.15121v2","created_at":"2026-07-05T11:51:32.375675+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.15121","created_at":"2026-07-05T11:51:32.375675+00:00"},{"alias_kind":"pith_short_12","alias_value":"FBF3MTBHSPI4","created_at":"2026-07-05T11:51:32.375675+00:00"},{"alias_kind":"pith_short_16","alias_value":"FBF3MTBHSPI4HVAQ","created_at":"2026-07-05T11:51:32.375675+00:00"},{"alias_kind":"pith_short_8","alias_value":"FBF3MTBH","created_at":"2026-07-05T11:51:32.375675+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FBF3MTBHSPI4HVAQLJ733QP7T6","json":"https://pith.science/pith/FBF3MTBHSPI4HVAQLJ733QP7T6.json","graph_json":"https://pith.science/api/pith-number/FBF3MTBHSPI4HVAQLJ733QP7T6/graph.json","events_json":"https://pith.science/api/pith-number/FBF3MTBHSPI4HVAQLJ733QP7T6/events.json","paper":"https://pith.science/paper/FBF3MTBH"},"agent_actions":{"view_html":"https://pith.science/pith/FBF3MTBHSPI4HVAQLJ733QP7T6","download_json":"https://pith.science/pith/FBF3MTBHSPI4HVAQLJ733QP7T6.json","view_paper":"https://pith.science/paper/FBF3MTBH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.15121&json=true","fetch_graph":"https://pith.science/api/pith-number/FBF3MTBHSPI4HVAQLJ733QP7T6/graph.json","fetch_events":"https://pith.science/api/pith-number/FBF3MTBHSPI4HVAQLJ733QP7T6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FBF3MTBHSPI4HVAQLJ733QP7T6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FBF3MTBHSPI4HVAQLJ733QP7T6/action/storage_attestation","attest_author":"https://pith.science/pith/FBF3MTBHSPI4HVAQLJ733QP7T6/action/author_attestation","sign_citation":"https://pith.science/pith/FBF3MTBHSPI4HVAQLJ733QP7T6/action/citation_signature","submit_replication":"https://pith.science/pith/FBF3MTBHSPI4HVAQLJ733QP7T6/action/replication_record"}},"created_at":"2026-07-05T11:51:32.375675+00:00","updated_at":"2026-07-05T11:51:32.375675+00:00"}