{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:X5AHL4BGWROMKOLRABP4RNAVED","short_pith_number":"pith:X5AHL4BG","canonical_record":{"source":{"id":"1811.12174","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2018-11-29T14:22:05Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"9f67b995559506165223c06e7ff4e9d595f6919cc2a06baf332abc1eb3065282","abstract_canon_sha256":"84dfe05461a9d054acc362521751c075b34e71d086731944f35163a8376b586c"},"schema_version":"1.0"},"canonical_sha256":"bf4075f026b45cc53971005fc8b41520d14719d2bda0a790bf2d5f87c1af6562","source":{"kind":"arxiv","id":"1811.12174","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1811.12174","created_at":"2026-05-17T23:59:34Z"},{"alias_kind":"arxiv_version","alias_value":"1811.12174v1","created_at":"2026-05-17T23:59:34Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1811.12174","created_at":"2026-05-17T23:59:34Z"},{"alias_kind":"pith_short_12","alias_value":"X5AHL4BGWROM","created_at":"2026-05-18T12:33:01Z"},{"alias_kind":"pith_short_16","alias_value":"X5AHL4BGWROMKOLR","created_at":"2026-05-18T12:33:01Z"},{"alias_kind":"pith_short_8","alias_value":"X5AHL4BG","created_at":"2026-05-18T12:33:01Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:X5AHL4BGWROMKOLRABP4RNAVED","target":"record","payload":{"canonical_record":{"source":{"id":"1811.12174","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2018-11-29T14:22:05Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"9f67b995559506165223c06e7ff4e9d595f6919cc2a06baf332abc1eb3065282","abstract_canon_sha256":"84dfe05461a9d054acc362521751c075b34e71d086731944f35163a8376b586c"},"schema_version":"1.0"},"canonical_sha256":"bf4075f026b45cc53971005fc8b41520d14719d2bda0a790bf2d5f87c1af6562","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:59:34.214339Z","signature_b64":"yIWFbWdLBBJ6Af0/6wYVOzw7hf4Uzh9NGTP+txirS3xJHUjHqmFLhdyJsV24Yd+vf9oNzgOMp05IVANpqVxxAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bf4075f026b45cc53971005fc8b41520d14719d2bda0a790bf2d5f87c1af6562","last_reissued_at":"2026-05-17T23:59:34.213597Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:59:34.213597Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1811.12174","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":"2026-05-17T23:59:34Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"VEugRnqIKhmbFSWIZtC5NhIHHgTn8mnmYM4rG9HJxDA7JIETRvMhXyeEYsugeLmBEKIWmXPessZk/LPJewjaDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-01T22:37:17.979511Z"},"content_sha256":"9066227cf0587b5538c425aa115510daa00c1008d8158e1d19a0e6dc693891cd","schema_version":"1.0","event_id":"sha256:9066227cf0587b5538c425aa115510daa00c1008d8158e1d19a0e6dc693891cd"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:X5AHL4BGWROMKOLRABP4RNAVED","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Data-parallel distributed training of very large models beyond GPU capacity","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.DC","authors_text":"Amit Juneja, Anar Yusifov, Bryant Nelson, Max Grossman, Minsik Cho, Samuel Matzek","submitted_at":"2018-11-29T14:22:05Z","abstract_excerpt":"GPUs have limited memory and it is difficult to train wide and/or deep models that cause the training process to go out of memory. It is shown in this paper how an open source tool called Large Model Support (LMS) can utilize a high bandwidth NVLink connection between CPUs and GPUs to accomplish training of deep convolutional networks. LMS performs tensor swapping between CPU memory and GPU memory such that only a minimal number of tensors required in a training step are kept in the GPU memory. It is also shown how LMS can be combined with an MPI based distributed deep learning module to train"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1811.12174","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":""},"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":"2026-05-17T23:59:34Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"c+iDijrApuZbV1SKqLe9S8D/UKW07S/SukwzGM0p7hMMX5MP00E8L+rrh+PGDt+gsMHXSe8+CcMYQacQKPh9AQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-01T22:37:17.979853Z"},"content_sha256":"a21566cc77a2e62edb44ee779d0b4fce936cd8f673d05152b326f695335bcd04","schema_version":"1.0","event_id":"sha256:a21566cc77a2e62edb44ee779d0b4fce936cd8f673d05152b326f695335bcd04"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/X5AHL4BGWROMKOLRABP4RNAVED/bundle.json","state_url":"https://pith.science/pith/X5AHL4BGWROMKOLRABP4RNAVED/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/X5AHL4BGWROMKOLRABP4RNAVED/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":"2026-06-01T22:37:17Z","links":{"resolver":"https://pith.science/pith/X5AHL4BGWROMKOLRABP4RNAVED","bundle":"https://pith.science/pith/X5AHL4BGWROMKOLRABP4RNAVED/bundle.json","state":"https://pith.science/pith/X5AHL4BGWROMKOLRABP4RNAVED/state.json","well_known_bundle":"https://pith.science/.well-known/pith/X5AHL4BGWROMKOLRABP4RNAVED/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:X5AHL4BGWROMKOLRABP4RNAVED","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":"84dfe05461a9d054acc362521751c075b34e71d086731944f35163a8376b586c","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2018-11-29T14:22:05Z","title_canon_sha256":"9f67b995559506165223c06e7ff4e9d595f6919cc2a06baf332abc1eb3065282"},"schema_version":"1.0","source":{"id":"1811.12174","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1811.12174","created_at":"2026-05-17T23:59:34Z"},{"alias_kind":"arxiv_version","alias_value":"1811.12174v1","created_at":"2026-05-17T23:59:34Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1811.12174","created_at":"2026-05-17T23:59:34Z"},{"alias_kind":"pith_short_12","alias_value":"X5AHL4BGWROM","created_at":"2026-05-18T12:33:01Z"},{"alias_kind":"pith_short_16","alias_value":"X5AHL4BGWROMKOLR","created_at":"2026-05-18T12:33:01Z"},{"alias_kind":"pith_short_8","alias_value":"X5AHL4BG","created_at":"2026-05-18T12:33:01Z"}],"graph_snapshots":[{"event_id":"sha256:a21566cc77a2e62edb44ee779d0b4fce936cd8f673d05152b326f695335bcd04","target":"graph","created_at":"2026-05-17T23:59:34Z","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":"GPUs have limited memory and it is difficult to train wide and/or deep models that cause the training process to go out of memory. It is shown in this paper how an open source tool called Large Model Support (LMS) can utilize a high bandwidth NVLink connection between CPUs and GPUs to accomplish training of deep convolutional networks. LMS performs tensor swapping between CPU memory and GPU memory such that only a minimal number of tensors required in a training step are kept in the GPU memory. It is also shown how LMS can be combined with an MPI based distributed deep learning module to train","authors_text":"Amit Juneja, Anar Yusifov, Bryant Nelson, Max Grossman, Minsik Cho, Samuel Matzek","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2018-11-29T14:22:05Z","title":"Data-parallel distributed training of very large models beyond GPU capacity"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1811.12174","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":"sha256:9066227cf0587b5538c425aa115510daa00c1008d8158e1d19a0e6dc693891cd","target":"record","created_at":"2026-05-17T23:59:34Z","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":"84dfe05461a9d054acc362521751c075b34e71d086731944f35163a8376b586c","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2018-11-29T14:22:05Z","title_canon_sha256":"9f67b995559506165223c06e7ff4e9d595f6919cc2a06baf332abc1eb3065282"},"schema_version":"1.0","source":{"id":"1811.12174","kind":"arxiv","version":1}},"canonical_sha256":"bf4075f026b45cc53971005fc8b41520d14719d2bda0a790bf2d5f87c1af6562","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"bf4075f026b45cc53971005fc8b41520d14719d2bda0a790bf2d5f87c1af6562","first_computed_at":"2026-05-17T23:59:34.213597Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:59:34.213597Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"yIWFbWdLBBJ6Af0/6wYVOzw7hf4Uzh9NGTP+txirS3xJHUjHqmFLhdyJsV24Yd+vf9oNzgOMp05IVANpqVxxAg==","signature_status":"signed_v1","signed_at":"2026-05-17T23:59:34.214339Z","signed_message":"canonical_sha256_bytes"},"source_id":"1811.12174","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9066227cf0587b5538c425aa115510daa00c1008d8158e1d19a0e6dc693891cd","sha256:a21566cc77a2e62edb44ee779d0b4fce936cd8f673d05152b326f695335bcd04"],"state_sha256":"4481125b690a53319744cfd705c3f35083b6b536167512359709feb81a329e52"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0v1ooyoxWReJZ8di0ekZMdybz915jzhX7cpPSrzvzrWUh9RF4f/LFh8fZ6DLebREEOToJODrT0Hcao83bQJYAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-01T22:37:17.981737Z","bundle_sha256":"5bcbe53c642f891baf66dfb725c395eca16a8b73b030d70848f2aa3c307653c2"}}