{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2020:53UIAZZKKOBTWPZ3GTEVCM3AEX","short_pith_number":"pith:53UIAZZK","canonical_record":{"source":{"id":"2005.13425","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2020-05-27T15:30:11Z","cross_cats_sorted":[],"title_canon_sha256":"bf2c5bc3ffb8186ba2ad3af077ae6ef97bbbb553f277d1f9311cca9f04275f1b","abstract_canon_sha256":"0988fedce7ef135a7a74a1c6768f7953cccf6c23b5ed4b368445df5da8c7863d"},"schema_version":"1.0"},"canonical_sha256":"eee880672a53833b3f3b34c951336025e668cb9030eeee20730a42c603571ec8","source":{"kind":"arxiv","id":"2005.13425","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2005.13425","created_at":"2026-07-05T01:06:15Z"},{"alias_kind":"arxiv_version","alias_value":"2005.13425v1","created_at":"2026-07-05T01:06:15Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2005.13425","created_at":"2026-07-05T01:06:15Z"},{"alias_kind":"pith_short_12","alias_value":"53UIAZZKKOBT","created_at":"2026-07-05T01:06:15Z"},{"alias_kind":"pith_short_16","alias_value":"53UIAZZKKOBTWPZ3","created_at":"2026-07-05T01:06:15Z"},{"alias_kind":"pith_short_8","alias_value":"53UIAZZK","created_at":"2026-07-05T01:06:15Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2020:53UIAZZKKOBTWPZ3GTEVCM3AEX","target":"record","payload":{"canonical_record":{"source":{"id":"2005.13425","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2020-05-27T15:30:11Z","cross_cats_sorted":[],"title_canon_sha256":"bf2c5bc3ffb8186ba2ad3af077ae6ef97bbbb553f277d1f9311cca9f04275f1b","abstract_canon_sha256":"0988fedce7ef135a7a74a1c6768f7953cccf6c23b5ed4b368445df5da8c7863d"},"schema_version":"1.0"},"canonical_sha256":"eee880672a53833b3f3b34c951336025e668cb9030eeee20730a42c603571ec8","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:06:15.766479Z","signature_b64":"nrzgthO8ZP6bTz6gkDVGkBLkSD9/lUqH0m7YSxa3a0Q6kw4SbCsrP0+Ga7EI2H1EfCob0UQugJigVfp4F+exBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"eee880672a53833b3f3b34c951336025e668cb9030eeee20730a42c603571ec8","last_reissued_at":"2026-07-05T01:06:15.766057Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:06:15.766057Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2005.13425","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T01:06:15Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"n44b9PfvEsHNbn1+vIqTBcU0p3DhdF/1TtD6/7cCjsiRLGO0xxl1tV9bXhog/GXmLRZaPjyqwaco5mmE1pj1DA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T06:30:34.386214Z"},"content_sha256":"0c3b731f41091cab5d32827ea2c82bcd9416f9d8ed7d94876fff4d0a5aef4369","schema_version":"1.0","event_id":"sha256:0c3b731f41091cab5d32827ea2c82bcd9416f9d8ed7d94876fff4d0a5aef4369"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2020:53UIAZZKKOBTWPZ3GTEVCM3AEX","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Optimization of Tensor-product Operations in Nekbone on GPUs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.DC","authors_text":"Artur Podobas, Martin Karp, Niclas Jansson, Philipp Schlatter, Stefano Markidis","submitted_at":"2020-05-27T15:30:11Z","abstract_excerpt":"In the CFD solver Nek5000, the computation is dominated by the evaluation of small tensor operations. Nekbone is a proxy app for Nek5000 and has previously been ported to GPUs with a mixed OpenACC and CUDA approach. In this work, we continue this effort and optimize the main tensor-product operation in Nekbone further. Our optimization is done in CUDA and uses a different, 2D, thread structure to make the computations layer by layer. This enables us to use loop unrolling as well as utilize registers and shared memory efficiently. Our implementation is then compared on both the Pascal and Volta"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2005.13425","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/2005.13425/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T01:06:15Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mGM3xqiR+G9ueGx5U+YPzY0tfYBCkHvAbDuO3MdlOdXdh8Off9pa5wmYbpzwKhEMIB5xZOTryXD7oKiKjGivAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T06:30:34.386607Z"},"content_sha256":"1573e933feed02c2f0c63c83877885dd1adfd051aeaa4870f643305627646c86","schema_version":"1.0","event_id":"sha256:1573e933feed02c2f0c63c83877885dd1adfd051aeaa4870f643305627646c86"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/53UIAZZKKOBTWPZ3GTEVCM3AEX/bundle.json","state_url":"https://pith.science/pith/53UIAZZKKOBTWPZ3GTEVCM3AEX/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/53UIAZZKKOBTWPZ3GTEVCM3AEX/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-07-07T06:30:34Z","links":{"resolver":"https://pith.science/pith/53UIAZZKKOBTWPZ3GTEVCM3AEX","bundle":"https://pith.science/pith/53UIAZZKKOBTWPZ3GTEVCM3AEX/bundle.json","state":"https://pith.science/pith/53UIAZZKKOBTWPZ3GTEVCM3AEX/state.json","well_known_bundle":"https://pith.science/.well-known/pith/53UIAZZKKOBTWPZ3GTEVCM3AEX/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2020:53UIAZZKKOBTWPZ3GTEVCM3AEX","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"0988fedce7ef135a7a74a1c6768f7953cccf6c23b5ed4b368445df5da8c7863d","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2020-05-27T15:30:11Z","title_canon_sha256":"bf2c5bc3ffb8186ba2ad3af077ae6ef97bbbb553f277d1f9311cca9f04275f1b"},"schema_version":"1.0","source":{"id":"2005.13425","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2005.13425","created_at":"2026-07-05T01:06:15Z"},{"alias_kind":"arxiv_version","alias_value":"2005.13425v1","created_at":"2026-07-05T01:06:15Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2005.13425","created_at":"2026-07-05T01:06:15Z"},{"alias_kind":"pith_short_12","alias_value":"53UIAZZKKOBT","created_at":"2026-07-05T01:06:15Z"},{"alias_kind":"pith_short_16","alias_value":"53UIAZZKKOBTWPZ3","created_at":"2026-07-05T01:06:15Z"},{"alias_kind":"pith_short_8","alias_value":"53UIAZZK","created_at":"2026-07-05T01:06:15Z"}],"graph_snapshots":[{"event_id":"sha256:1573e933feed02c2f0c63c83877885dd1adfd051aeaa4870f643305627646c86","target":"graph","created_at":"2026-07-05T01:06:15Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2005.13425/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"In the CFD solver Nek5000, the computation is dominated by the evaluation of small tensor operations. Nekbone is a proxy app for Nek5000 and has previously been ported to GPUs with a mixed OpenACC and CUDA approach. In this work, we continue this effort and optimize the main tensor-product operation in Nekbone further. Our optimization is done in CUDA and uses a different, 2D, thread structure to make the computations layer by layer. This enables us to use loop unrolling as well as utilize registers and shared memory efficiently. Our implementation is then compared on both the Pascal and Volta","authors_text":"Artur Podobas, Martin Karp, Niclas Jansson, Philipp Schlatter, Stefano Markidis","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2020-05-27T15:30:11Z","title":"Optimization of Tensor-product Operations in Nekbone on GPUs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2005.13425","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:0c3b731f41091cab5d32827ea2c82bcd9416f9d8ed7d94876fff4d0a5aef4369","target":"record","created_at":"2026-07-05T01:06:15Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"0988fedce7ef135a7a74a1c6768f7953cccf6c23b5ed4b368445df5da8c7863d","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2020-05-27T15:30:11Z","title_canon_sha256":"bf2c5bc3ffb8186ba2ad3af077ae6ef97bbbb553f277d1f9311cca9f04275f1b"},"schema_version":"1.0","source":{"id":"2005.13425","kind":"arxiv","version":1}},"canonical_sha256":"eee880672a53833b3f3b34c951336025e668cb9030eeee20730a42c603571ec8","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"eee880672a53833b3f3b34c951336025e668cb9030eeee20730a42c603571ec8","first_computed_at":"2026-07-05T01:06:15.766057Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T01:06:15.766057Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"nrzgthO8ZP6bTz6gkDVGkBLkSD9/lUqH0m7YSxa3a0Q6kw4SbCsrP0+Ga7EI2H1EfCob0UQugJigVfp4F+exBA==","signature_status":"signed_v1","signed_at":"2026-07-05T01:06:15.766479Z","signed_message":"canonical_sha256_bytes"},"source_id":"2005.13425","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0c3b731f41091cab5d32827ea2c82bcd9416f9d8ed7d94876fff4d0a5aef4369","sha256:1573e933feed02c2f0c63c83877885dd1adfd051aeaa4870f643305627646c86"],"state_sha256":"d562be0628926e6f0b3f33598102924afc25fdce1a119b049f1a49a0a9cb9f8b"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HLro5sA1CBRfcoywJuRQvwhY3DRPvdPRfgR7yFLcSmo8LkaDKw2qhOr90oTVZO6C90Nh7rOsK8bmcBfrXQyMBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-07T06:30:34.388946Z","bundle_sha256":"fce810201740341d753c01d900455fbc3b04575ec9a7ef204e6b4055641c1af7"}}