{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:GO43K6A5MHLNBUADUVZ3AKFBNI","short_pith_number":"pith:GO43K6A5","schema_version":"1.0","canonical_sha256":"33b9b5781d61d6d0d003a573b028a16a34edfc01fff0c37e4df8bf623276f1f9","source":{"kind":"arxiv","id":"2111.04007","version":2},"attestation_state":"computed","paper":{"title":"Varuna: Scalable, Low-cost Training of Massive Deep Learning Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.DC","authors_text":"Muthian Sivathanu, Nipun Kwatra, Nitika Saran, Ramachandran Ramjee, Sanjith Athlur","submitted_at":"2021-11-07T05:33:48Z","abstract_excerpt":"Systems for training massive deep learning models (billions of parameters) today assume and require specialized \"hyper-clusters\": hundreds or thousands of GPUs wired with specialized high-bandwidth interconnects such as NV-Link and Infiniband. Besides being expensive, such dependence on hyper-clusters and custom high-speed inter-connects limits the size of such clusters, creating (a) scalability limits on job parallelism; (b) resource fragmentation across hyper-clusters.\n  In this paper, we present Varuna, a new system that enables training massive deep learning models on commodity networking."},"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":"2111.04007","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2021-11-07T05:33:48Z","cross_cats_sorted":[],"title_canon_sha256":"ca8dbc2362a585fc0c3bebb63bd5e936bf9ce9fc18f5869ad68d5a8e3dcd5569","abstract_canon_sha256":"50aef61e172885efe463dd355a10ae03764ce0098d53b6b65617151c225ca066"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:31:32.238357Z","signature_b64":"iA1jySv92T6Y2Euu/zMXcNwELs0wXQkn7R3tfpOfKTII43f4h1pvQ3ScoaZASMFYwhBaEOmWB5iIkRqou3ITBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"33b9b5781d61d6d0d003a573b028a16a34edfc01fff0c37e4df8bf623276f1f9","last_reissued_at":"2026-07-05T03:31:32.237888Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:31:32.237888Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Varuna: Scalable, Low-cost Training of Massive Deep Learning Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.DC","authors_text":"Muthian Sivathanu, Nipun Kwatra, Nitika Saran, Ramachandran Ramjee, Sanjith Athlur","submitted_at":"2021-11-07T05:33:48Z","abstract_excerpt":"Systems for training massive deep learning models (billions of parameters) today assume and require specialized \"hyper-clusters\": hundreds or thousands of GPUs wired with specialized high-bandwidth interconnects such as NV-Link and Infiniband. Besides being expensive, such dependence on hyper-clusters and custom high-speed inter-connects limits the size of such clusters, creating (a) scalability limits on job parallelism; (b) resource fragmentation across hyper-clusters.\n  In this paper, we present Varuna, a new system that enables training massive deep learning models on commodity networking."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2111.04007","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/2111.04007/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":"2111.04007","created_at":"2026-07-05T03:31:32.237949+00:00"},{"alias_kind":"arxiv_version","alias_value":"2111.04007v2","created_at":"2026-07-05T03:31:32.237949+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2111.04007","created_at":"2026-07-05T03:31:32.237949+00:00"},{"alias_kind":"pith_short_12","alias_value":"GO43K6A5MHLN","created_at":"2026-07-05T03:31:32.237949+00:00"},{"alias_kind":"pith_short_16","alias_value":"GO43K6A5MHLNBUAD","created_at":"2026-07-05T03:31:32.237949+00:00"},{"alias_kind":"pith_short_8","alias_value":"GO43K6A5","created_at":"2026-07-05T03:31:32.237949+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/GO43K6A5MHLNBUADUVZ3AKFBNI","json":"https://pith.science/pith/GO43K6A5MHLNBUADUVZ3AKFBNI.json","graph_json":"https://pith.science/api/pith-number/GO43K6A5MHLNBUADUVZ3AKFBNI/graph.json","events_json":"https://pith.science/api/pith-number/GO43K6A5MHLNBUADUVZ3AKFBNI/events.json","paper":"https://pith.science/paper/GO43K6A5"},"agent_actions":{"view_html":"https://pith.science/pith/GO43K6A5MHLNBUADUVZ3AKFBNI","download_json":"https://pith.science/pith/GO43K6A5MHLNBUADUVZ3AKFBNI.json","view_paper":"https://pith.science/paper/GO43K6A5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2111.04007&json=true","fetch_graph":"https://pith.science/api/pith-number/GO43K6A5MHLNBUADUVZ3AKFBNI/graph.json","fetch_events":"https://pith.science/api/pith-number/GO43K6A5MHLNBUADUVZ3AKFBNI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GO43K6A5MHLNBUADUVZ3AKFBNI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GO43K6A5MHLNBUADUVZ3AKFBNI/action/storage_attestation","attest_author":"https://pith.science/pith/GO43K6A5MHLNBUADUVZ3AKFBNI/action/author_attestation","sign_citation":"https://pith.science/pith/GO43K6A5MHLNBUADUVZ3AKFBNI/action/citation_signature","submit_replication":"https://pith.science/pith/GO43K6A5MHLNBUADUVZ3AKFBNI/action/replication_record"}},"created_at":"2026-07-05T03:31:32.237949+00:00","updated_at":"2026-07-05T03:31:32.237949+00:00"}