{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:46OHPU4L24WTERPE7MV2M5M6HT","short_pith_number":"pith:46OHPU4L","schema_version":"1.0","canonical_sha256":"e79c77d38bd72d3245e4fb2ba6759e3cf44b43f9f389cd2e1e80cfba4e82d24a","source":{"kind":"arxiv","id":"1810.11112","version":1},"attestation_state":"computed","paper":{"title":"Scalable Distributed DNN Training using TensorFlow and CUDA-Aware MPI: Characterization, Designs, and Performance Evaluation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.DC","authors_text":"Ammar Ahmad Awan, Ching-Hsiang Chu, Dhabaleswar K. Panda, Hari Subramoni, Jeroen Bedorf","submitted_at":"2018-10-25T21:25:26Z","abstract_excerpt":"TensorFlow has been the most widely adopted Machine/Deep Learning framework. However, little exists in the literature that provides a thorough understanding of the capabilities which TensorFlow offers for the distributed training of large ML/DL models that need computation and communication at scale. Most commonly used distributed training approaches for TF can be categorized as follows: 1) Google Remote Procedure Call (gRPC), 2) gRPC+X: X=(InfiniBand Verbs, Message Passing Interface, and GPUDirect RDMA), and 3) No-gRPC: Baidu Allreduce with MPI, Horovod with MPI, and Horovod with NVIDIA NCCL."},"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":"1810.11112","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2018-10-25T21:25:26Z","cross_cats_sorted":[],"title_canon_sha256":"2c9b579fb424972653c1d679a082d1823af2886cf01ae9c05bbe132f73616302","abstract_canon_sha256":"b608c8bbc37b541647f4cc4627784ec15507bcbc3ff340fdae47cd6e92329761"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:18:49.904815Z","signature_b64":"xiw4P0iGYiFYBaAFKCCXEkBEca3MLXkBQlgGiqDdo1hLWehDqvjI1FRny+vNEog/zonCKOrVextoz3+AidYQAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e79c77d38bd72d3245e4fb2ba6759e3cf44b43f9f389cd2e1e80cfba4e82d24a","last_reissued_at":"2026-07-05T00:18:49.904354Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:18:49.904354Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Scalable Distributed DNN Training using TensorFlow and CUDA-Aware MPI: Characterization, Designs, and Performance Evaluation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.DC","authors_text":"Ammar Ahmad Awan, Ching-Hsiang Chu, Dhabaleswar K. Panda, Hari Subramoni, Jeroen Bedorf","submitted_at":"2018-10-25T21:25:26Z","abstract_excerpt":"TensorFlow has been the most widely adopted Machine/Deep Learning framework. However, little exists in the literature that provides a thorough understanding of the capabilities which TensorFlow offers for the distributed training of large ML/DL models that need computation and communication at scale. Most commonly used distributed training approaches for TF can be categorized as follows: 1) Google Remote Procedure Call (gRPC), 2) gRPC+X: X=(InfiniBand Verbs, Message Passing Interface, and GPUDirect RDMA), and 3) No-gRPC: Baidu Allreduce with MPI, Horovod with MPI, and Horovod with NVIDIA NCCL."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1810.11112","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/1810.11112/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":"1810.11112","created_at":"2026-07-05T00:18:49.904410+00:00"},{"alias_kind":"arxiv_version","alias_value":"1810.11112v1","created_at":"2026-07-05T00:18:49.904410+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1810.11112","created_at":"2026-07-05T00:18:49.904410+00:00"},{"alias_kind":"pith_short_12","alias_value":"46OHPU4L24WT","created_at":"2026-07-05T00:18:49.904410+00:00"},{"alias_kind":"pith_short_16","alias_value":"46OHPU4L24WTERPE","created_at":"2026-07-05T00:18:49.904410+00:00"},{"alias_kind":"pith_short_8","alias_value":"46OHPU4L","created_at":"2026-07-05T00:18:49.904410+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/46OHPU4L24WTERPE7MV2M5M6HT","json":"https://pith.science/pith/46OHPU4L24WTERPE7MV2M5M6HT.json","graph_json":"https://pith.science/api/pith-number/46OHPU4L24WTERPE7MV2M5M6HT/graph.json","events_json":"https://pith.science/api/pith-number/46OHPU4L24WTERPE7MV2M5M6HT/events.json","paper":"https://pith.science/paper/46OHPU4L"},"agent_actions":{"view_html":"https://pith.science/pith/46OHPU4L24WTERPE7MV2M5M6HT","download_json":"https://pith.science/pith/46OHPU4L24WTERPE7MV2M5M6HT.json","view_paper":"https://pith.science/paper/46OHPU4L","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1810.11112&json=true","fetch_graph":"https://pith.science/api/pith-number/46OHPU4L24WTERPE7MV2M5M6HT/graph.json","fetch_events":"https://pith.science/api/pith-number/46OHPU4L24WTERPE7MV2M5M6HT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/46OHPU4L24WTERPE7MV2M5M6HT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/46OHPU4L24WTERPE7MV2M5M6HT/action/storage_attestation","attest_author":"https://pith.science/pith/46OHPU4L24WTERPE7MV2M5M6HT/action/author_attestation","sign_citation":"https://pith.science/pith/46OHPU4L24WTERPE7MV2M5M6HT/action/citation_signature","submit_replication":"https://pith.science/pith/46OHPU4L24WTERPE7MV2M5M6HT/action/replication_record"}},"created_at":"2026-07-05T00:18:49.904410+00:00","updated_at":"2026-07-05T00:18:49.904410+00:00"}