{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:SDUSFBJVF5NR7RR4RZORWNLRUH","short_pith_number":"pith:SDUSFBJV","schema_version":"1.0","canonical_sha256":"90e92285352f5b1fc63c8e5d1b3571a1e6e0540bc0845351db9d20012f4a6ec0","source":{"kind":"arxiv","id":"2503.22643","version":2},"attestation_state":"computed","paper":{"title":"Hiding Latencies in Network-Based Image Loading for Deep Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.DC","authors_text":"Francesco Versaci, Giovanni Busonera","submitted_at":"2025-03-28T17:31:03Z","abstract_excerpt":"In the last decades, the computational power of GPUs has grown exponentially, allowing current deep learning (DL) applications to handle increasingly large amounts of data at a progressively higher throughput. However, network and storage latencies cannot decrease at a similar pace due to physical constraints, leading to data stalls, and creating a bottleneck for DL tasks. Additionally, managing vast quantities of data and their associated metadata has proven challenging, hampering and slowing the productivity of data scientists. Moreover, existing data loaders have limited network support, ne"},"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":"2503.22643","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2025-03-28T17:31:03Z","cross_cats_sorted":[],"title_canon_sha256":"0c726984543096a97df733a4df35468e4eaca70387a81df92ec0af164ea0e82d","abstract_canon_sha256":"69ad40a501a93dd6c1cd3835bc85ca58b98fbbc2ad606be3dd1eb5f465c64308"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:05:15.423101Z","signature_b64":"NQgmEH2Zb5mB+j/Cmm1RZZiGH9UqleWvqrl5WlExiIgc04wq/kaNcFoBYcaiir0Etj5WBCTSq7x3p/hKVEYXBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"90e92285352f5b1fc63c8e5d1b3571a1e6e0540bc0845351db9d20012f4a6ec0","last_reissued_at":"2026-07-05T12:05:15.422487Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:05:15.422487Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Hiding Latencies in Network-Based Image Loading for Deep Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.DC","authors_text":"Francesco Versaci, Giovanni Busonera","submitted_at":"2025-03-28T17:31:03Z","abstract_excerpt":"In the last decades, the computational power of GPUs has grown exponentially, allowing current deep learning (DL) applications to handle increasingly large amounts of data at a progressively higher throughput. However, network and storage latencies cannot decrease at a similar pace due to physical constraints, leading to data stalls, and creating a bottleneck for DL tasks. Additionally, managing vast quantities of data and their associated metadata has proven challenging, hampering and slowing the productivity of data scientists. Moreover, existing data loaders have limited network support, ne"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.22643","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/2503.22643/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":"2503.22643","created_at":"2026-07-05T12:05:15.422564+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.22643v2","created_at":"2026-07-05T12:05:15.422564+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.22643","created_at":"2026-07-05T12:05:15.422564+00:00"},{"alias_kind":"pith_short_12","alias_value":"SDUSFBJVF5NR","created_at":"2026-07-05T12:05:15.422564+00:00"},{"alias_kind":"pith_short_16","alias_value":"SDUSFBJVF5NR7RR4","created_at":"2026-07-05T12:05:15.422564+00:00"},{"alias_kind":"pith_short_8","alias_value":"SDUSFBJV","created_at":"2026-07-05T12:05:15.422564+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2508.11035","citing_title":"EMLIO: Minimizing I/O Latency and Energy Consumption for Large-Scale AI Training","ref_index":49,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SDUSFBJVF5NR7RR4RZORWNLRUH","json":"https://pith.science/pith/SDUSFBJVF5NR7RR4RZORWNLRUH.json","graph_json":"https://pith.science/api/pith-number/SDUSFBJVF5NR7RR4RZORWNLRUH/graph.json","events_json":"https://pith.science/api/pith-number/SDUSFBJVF5NR7RR4RZORWNLRUH/events.json","paper":"https://pith.science/paper/SDUSFBJV"},"agent_actions":{"view_html":"https://pith.science/pith/SDUSFBJVF5NR7RR4RZORWNLRUH","download_json":"https://pith.science/pith/SDUSFBJVF5NR7RR4RZORWNLRUH.json","view_paper":"https://pith.science/paper/SDUSFBJV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.22643&json=true","fetch_graph":"https://pith.science/api/pith-number/SDUSFBJVF5NR7RR4RZORWNLRUH/graph.json","fetch_events":"https://pith.science/api/pith-number/SDUSFBJVF5NR7RR4RZORWNLRUH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SDUSFBJVF5NR7RR4RZORWNLRUH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SDUSFBJVF5NR7RR4RZORWNLRUH/action/storage_attestation","attest_author":"https://pith.science/pith/SDUSFBJVF5NR7RR4RZORWNLRUH/action/author_attestation","sign_citation":"https://pith.science/pith/SDUSFBJVF5NR7RR4RZORWNLRUH/action/citation_signature","submit_replication":"https://pith.science/pith/SDUSFBJVF5NR7RR4RZORWNLRUH/action/replication_record"}},"created_at":"2026-07-05T12:05:15.422564+00:00","updated_at":"2026-07-05T12:05:15.422564+00:00"}