{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2013:KKQIKKBGKXAONRNFCY6GXVJGIS","short_pith_number":"pith:KKQIKKBG","canonical_record":{"source":{"id":"1305.1183","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2013-05-06T13:32:22Z","cross_cats_sorted":[],"title_canon_sha256":"546ba3c4dd68270de3adaf9f768d2e0b1f99af64e9d290ecef11d6dc4b889d86","abstract_canon_sha256":"b7395ea44f35e3a745de1aeca5ef332459870fca19e1fd3fac7d638846644a8c"},"schema_version":"1.0"},"canonical_sha256":"52a085282655c0e6c5a5163c6bd52644b8f118528511683efac6c64279166eea","source":{"kind":"arxiv","id":"1305.1183","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1305.1183","created_at":"2026-05-18T00:35:36Z"},{"alias_kind":"arxiv_version","alias_value":"1305.1183v2","created_at":"2026-05-18T00:35:36Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1305.1183","created_at":"2026-05-18T00:35:36Z"},{"alias_kind":"pith_short_12","alias_value":"KKQIKKBGKXAO","created_at":"2026-05-18T12:27:51Z"},{"alias_kind":"pith_short_16","alias_value":"KKQIKKBGKXAONRNF","created_at":"2026-05-18T12:27:51Z"},{"alias_kind":"pith_short_8","alias_value":"KKQIKKBG","created_at":"2026-05-18T12:27:51Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2013:KKQIKKBGKXAONRNFCY6GXVJGIS","target":"record","payload":{"canonical_record":{"source":{"id":"1305.1183","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2013-05-06T13:32:22Z","cross_cats_sorted":[],"title_canon_sha256":"546ba3c4dd68270de3adaf9f768d2e0b1f99af64e9d290ecef11d6dc4b889d86","abstract_canon_sha256":"b7395ea44f35e3a745de1aeca5ef332459870fca19e1fd3fac7d638846644a8c"},"schema_version":"1.0"},"canonical_sha256":"52a085282655c0e6c5a5163c6bd52644b8f118528511683efac6c64279166eea","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:35:36.209541Z","signature_b64":"BuMaFgYJ41LOfSqm5ZfdoLVgyIxJi42NIrpR2pEyaL7Cg/w8B5/Aph0l0YFlVF0AE4AIeAfE/aVzHO08nU8aCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"52a085282655c0e6c5a5163c6bd52644b8f118528511683efac6c64279166eea","last_reissued_at":"2026-05-18T00:35:36.208930Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:35:36.208930Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1305.1183","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:35:36Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xhkfTV5TmMlgOoq8YLhVzYs60sH4/QSFFRqxTlVw60y3YgI9LzD42SCt1naegYZLWoEcaWTqmsGpUUhvCU2ECQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-10T18:09:44.990174Z"},"content_sha256":"a8259311e07ba624af9a90528a8e6d6132afe734493231900652ba0f364659b1","schema_version":"1.0","event_id":"sha256:a8259311e07ba624af9a90528a8e6d6132afe734493231900652ba0f364659b1"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2013:KKQIKKBGKXAONRNFCY6GXVJGIS","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Optimizing CUDA Code By Kernel Fusion---Application on BLAS","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.DC","authors_text":"J. Filipovi\\v{c}, J. Fousek, L. Matyska, M. Madzin","submitted_at":"2013-05-06T13:32:22Z","abstract_excerpt":"Modern GPUs are able to perform significantly more arithmetic operations than transfers of a single word to or from global memory. Hence, many GPU kernels are limited by memory bandwidth and cannot exploit the arithmetic power of GPUs. However, the memory locality can be often improved by kernel fusion when a sequence of kernels is executed and some kernels in this sequence share data.\n  In this paper, we show how kernels performing map, reduce or their nested combinations can be fused automatically by our source-to-source compiler. To demonstrate the usability of the compiler, we have impleme"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1305.1183","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":""},"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":"2026-05-18T00:35:36Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"85HcjD4EopEGbnnamHKknWG/Rg5UxLhTaK1z0vYcJi9fwO50W0agE6YiKC14dg8/t5Dt6V1XB82ZOT8BRaYfDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-10T18:09:44.990551Z"},"content_sha256":"cca979ad333aafb86fb8c4bc4be5695fb778f2fa5d3bcd95d38c1a5ba1fa08f2","schema_version":"1.0","event_id":"sha256:cca979ad333aafb86fb8c4bc4be5695fb778f2fa5d3bcd95d38c1a5ba1fa08f2"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/KKQIKKBGKXAONRNFCY6GXVJGIS/bundle.json","state_url":"https://pith.science/pith/KKQIKKBGKXAONRNFCY6GXVJGIS/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/KKQIKKBGKXAONRNFCY6GXVJGIS/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":"2026-06-10T18:09:44Z","links":{"resolver":"https://pith.science/pith/KKQIKKBGKXAONRNFCY6GXVJGIS","bundle":"https://pith.science/pith/KKQIKKBGKXAONRNFCY6GXVJGIS/bundle.json","state":"https://pith.science/pith/KKQIKKBGKXAONRNFCY6GXVJGIS/state.json","well_known_bundle":"https://pith.science/.well-known/pith/KKQIKKBGKXAONRNFCY6GXVJGIS/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2013:KKQIKKBGKXAONRNFCY6GXVJGIS","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":"b7395ea44f35e3a745de1aeca5ef332459870fca19e1fd3fac7d638846644a8c","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2013-05-06T13:32:22Z","title_canon_sha256":"546ba3c4dd68270de3adaf9f768d2e0b1f99af64e9d290ecef11d6dc4b889d86"},"schema_version":"1.0","source":{"id":"1305.1183","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1305.1183","created_at":"2026-05-18T00:35:36Z"},{"alias_kind":"arxiv_version","alias_value":"1305.1183v2","created_at":"2026-05-18T00:35:36Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1305.1183","created_at":"2026-05-18T00:35:36Z"},{"alias_kind":"pith_short_12","alias_value":"KKQIKKBGKXAO","created_at":"2026-05-18T12:27:51Z"},{"alias_kind":"pith_short_16","alias_value":"KKQIKKBGKXAONRNF","created_at":"2026-05-18T12:27:51Z"},{"alias_kind":"pith_short_8","alias_value":"KKQIKKBG","created_at":"2026-05-18T12:27:51Z"}],"graph_snapshots":[{"event_id":"sha256:cca979ad333aafb86fb8c4bc4be5695fb778f2fa5d3bcd95d38c1a5ba1fa08f2","target":"graph","created_at":"2026-05-18T00:35:36Z","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"},"paper":{"abstract_excerpt":"Modern GPUs are able to perform significantly more arithmetic operations than transfers of a single word to or from global memory. Hence, many GPU kernels are limited by memory bandwidth and cannot exploit the arithmetic power of GPUs. However, the memory locality can be often improved by kernel fusion when a sequence of kernels is executed and some kernels in this sequence share data.\n  In this paper, we show how kernels performing map, reduce or their nested combinations can be fused automatically by our source-to-source compiler. To demonstrate the usability of the compiler, we have impleme","authors_text":"J. Filipovi\\v{c}, J. Fousek, L. Matyska, M. Madzin","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2013-05-06T13:32:22Z","title":"Optimizing CUDA Code By Kernel Fusion---Application on BLAS"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1305.1183","kind":"arxiv","version":2},"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:a8259311e07ba624af9a90528a8e6d6132afe734493231900652ba0f364659b1","target":"record","created_at":"2026-05-18T00:35:36Z","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":"b7395ea44f35e3a745de1aeca5ef332459870fca19e1fd3fac7d638846644a8c","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2013-05-06T13:32:22Z","title_canon_sha256":"546ba3c4dd68270de3adaf9f768d2e0b1f99af64e9d290ecef11d6dc4b889d86"},"schema_version":"1.0","source":{"id":"1305.1183","kind":"arxiv","version":2}},"canonical_sha256":"52a085282655c0e6c5a5163c6bd52644b8f118528511683efac6c64279166eea","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"52a085282655c0e6c5a5163c6bd52644b8f118528511683efac6c64279166eea","first_computed_at":"2026-05-18T00:35:36.208930Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:35:36.208930Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"BuMaFgYJ41LOfSqm5ZfdoLVgyIxJi42NIrpR2pEyaL7Cg/w8B5/Aph0l0YFlVF0AE4AIeAfE/aVzHO08nU8aCQ==","signature_status":"signed_v1","signed_at":"2026-05-18T00:35:36.209541Z","signed_message":"canonical_sha256_bytes"},"source_id":"1305.1183","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a8259311e07ba624af9a90528a8e6d6132afe734493231900652ba0f364659b1","sha256:cca979ad333aafb86fb8c4bc4be5695fb778f2fa5d3bcd95d38c1a5ba1fa08f2"],"state_sha256":"3cd6ff4b2cab2df869b664a3b584a40d07b9d65dca3f83f8c59a27817df5cf4d"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0e35MhRfVZKe4gZtZ/bj0eyLJiDCFjMIFIZZyoHUd7jdBXjICgE8au/qEWuRz0LL7P9R0MZhbBwkJHAKVlNUBw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-10T18:09:44.992829Z","bundle_sha256":"3b2136eaa6032f0b4fe64197fda173db74dfb1feb1ab5097b6c72aa29e426955"}}