{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:5SZWLW4O4LL6TUMH7CZ75CR6UC","short_pith_number":"pith:5SZWLW4O","canonical_record":{"source":{"id":"2506.08132","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.NI","submitted_at":"2025-06-09T18:35:18Z","cross_cats_sorted":[],"title_canon_sha256":"5b53f5660b8ef7ad6d2dac001b36e86ded41415338a856e445de38f567b74fc2","abstract_canon_sha256":"6d7249dffda162e91d121c3074ec9ee7acef263bc660e55e848a94c008210135"},"schema_version":"1.0"},"canonical_sha256":"ecb365db8ee2d7e9d187f8b3fe8a3ea0abc503cc831f29e997193293b7229264","source":{"kind":"arxiv","id":"2506.08132","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.08132","created_at":"2026-07-05T11:18:45Z"},{"alias_kind":"arxiv_version","alias_value":"2506.08132v1","created_at":"2026-07-05T11:18:45Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.08132","created_at":"2026-07-05T11:18:45Z"},{"alias_kind":"pith_short_12","alias_value":"5SZWLW4O4LL6","created_at":"2026-07-05T11:18:45Z"},{"alias_kind":"pith_short_16","alias_value":"5SZWLW4O4LL6TUMH","created_at":"2026-07-05T11:18:45Z"},{"alias_kind":"pith_short_8","alias_value":"5SZWLW4O","created_at":"2026-07-05T11:18:45Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:5SZWLW4O4LL6TUMH7CZ75CR6UC","target":"record","payload":{"canonical_record":{"source":{"id":"2506.08132","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.NI","submitted_at":"2025-06-09T18:35:18Z","cross_cats_sorted":[],"title_canon_sha256":"5b53f5660b8ef7ad6d2dac001b36e86ded41415338a856e445de38f567b74fc2","abstract_canon_sha256":"6d7249dffda162e91d121c3074ec9ee7acef263bc660e55e848a94c008210135"},"schema_version":"1.0"},"canonical_sha256":"ecb365db8ee2d7e9d187f8b3fe8a3ea0abc503cc831f29e997193293b7229264","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:18:45.136755Z","signature_b64":"3QX23POwUbnlkwZZJnO3Ua0EKPvSEKMcWk4OzUZ3bbOpprwyc/z/OtRx2ziNY0woXqJfN0Ak7MfNhaZPaudFBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ecb365db8ee2d7e9d187f8b3fe8a3ea0abc503cc831f29e997193293b7229264","last_reissued_at":"2026-07-05T11:18:45.136272Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:18:45.136272Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.08132","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":"2026-07-05T11:18:45Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HQUaVcwGytUewk2Tty0jX6ZYpf7ARLLZNZDno5YBwTBKhh4ocMqyEAycFLtFuglf08tiGllZiyZydb1vxCVLAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T06:12:17.445839Z"},"content_sha256":"d9b2e119f8279cb7b1b8efa0b8339655cb29fd48944cf897cd03eaa3aa0a075d","schema_version":"1.0","event_id":"sha256:d9b2e119f8279cb7b1b8efa0b8339655cb29fd48944cf897cd03eaa3aa0a075d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:5SZWLW4O4LL6TUMH7CZ75CR6UC","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Congestion-Aware Path Selection for Load Balancing in AI Clusters","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.NI","authors_text":"Erfan Nosrati, Majid Ghaderi","submitted_at":"2025-06-09T18:35:18Z","abstract_excerpt":"Fast training of large machine learning models requires distributed training on AI clusters consisting of thousands of GPUs. The efficiency of distributed training crucially depends on the efficiency of the network interconnecting GPUs in the cluster. These networks are commonly built using RDMA following a Clos-like datacenter topology. To efficiently utilize the network bandwidth, load balancing is employed to distribute traffic across multiple redundant paths. While there exists numerous techniques for load-balancing in traditional datacenters, these are often either optimized for TCP traff"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.08132","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/2506.08132/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:18:45Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"pPUfQlSU5cDw1QOOOiCdV5gHWcm0v+A2bZww7R7B8/5v5ZI+8PgDlPxjqGKrTq/oTz1uuUq+brDllQJDZQAIBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T06:12:17.446213Z"},"content_sha256":"11cde9e0576681cf493d8278c2111acf7309b797e3e317c5830e9255031b2caf","schema_version":"1.0","event_id":"sha256:11cde9e0576681cf493d8278c2111acf7309b797e3e317c5830e9255031b2caf"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/5SZWLW4O4LL6TUMH7CZ75CR6UC/bundle.json","state_url":"https://pith.science/pith/5SZWLW4O4LL6TUMH7CZ75CR6UC/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/5SZWLW4O4LL6TUMH7CZ75CR6UC/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":"2026-08-18T06:12:17Z","links":{"resolver":"https://pith.science/pith/5SZWLW4O4LL6TUMH7CZ75CR6UC","bundle":"https://pith.science/pith/5SZWLW4O4LL6TUMH7CZ75CR6UC/bundle.json","state":"https://pith.science/pith/5SZWLW4O4LL6TUMH7CZ75CR6UC/state.json","well_known_bundle":"https://pith.science/.well-known/pith/5SZWLW4O4LL6TUMH7CZ75CR6UC/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:5SZWLW4O4LL6TUMH7CZ75CR6UC","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":"6d7249dffda162e91d121c3074ec9ee7acef263bc660e55e848a94c008210135","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.NI","submitted_at":"2025-06-09T18:35:18Z","title_canon_sha256":"5b53f5660b8ef7ad6d2dac001b36e86ded41415338a856e445de38f567b74fc2"},"schema_version":"1.0","source":{"id":"2506.08132","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.08132","created_at":"2026-07-05T11:18:45Z"},{"alias_kind":"arxiv_version","alias_value":"2506.08132v1","created_at":"2026-07-05T11:18:45Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.08132","created_at":"2026-07-05T11:18:45Z"},{"alias_kind":"pith_short_12","alias_value":"5SZWLW4O4LL6","created_at":"2026-07-05T11:18:45Z"},{"alias_kind":"pith_short_16","alias_value":"5SZWLW4O4LL6TUMH","created_at":"2026-07-05T11:18:45Z"},{"alias_kind":"pith_short_8","alias_value":"5SZWLW4O","created_at":"2026-07-05T11:18:45Z"}],"graph_snapshots":[{"event_id":"sha256:11cde9e0576681cf493d8278c2111acf7309b797e3e317c5830e9255031b2caf","target":"graph","created_at":"2026-07-05T11:18:45Z","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"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2506.08132/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Fast training of large machine learning models requires distributed training on AI clusters consisting of thousands of GPUs. The efficiency of distributed training crucially depends on the efficiency of the network interconnecting GPUs in the cluster. These networks are commonly built using RDMA following a Clos-like datacenter topology. To efficiently utilize the network bandwidth, load balancing is employed to distribute traffic across multiple redundant paths. While there exists numerous techniques for load-balancing in traditional datacenters, these are often either optimized for TCP traff","authors_text":"Erfan Nosrati, Majid Ghaderi","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.NI","submitted_at":"2025-06-09T18:35:18Z","title":"Congestion-Aware Path Selection for Load Balancing in AI Clusters"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.08132","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:d9b2e119f8279cb7b1b8efa0b8339655cb29fd48944cf897cd03eaa3aa0a075d","target":"record","created_at":"2026-07-05T11:18:45Z","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":"6d7249dffda162e91d121c3074ec9ee7acef263bc660e55e848a94c008210135","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.NI","submitted_at":"2025-06-09T18:35:18Z","title_canon_sha256":"5b53f5660b8ef7ad6d2dac001b36e86ded41415338a856e445de38f567b74fc2"},"schema_version":"1.0","source":{"id":"2506.08132","kind":"arxiv","version":1}},"canonical_sha256":"ecb365db8ee2d7e9d187f8b3fe8a3ea0abc503cc831f29e997193293b7229264","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ecb365db8ee2d7e9d187f8b3fe8a3ea0abc503cc831f29e997193293b7229264","first_computed_at":"2026-07-05T11:18:45.136272Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:18:45.136272Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"3QX23POwUbnlkwZZJnO3Ua0EKPvSEKMcWk4OzUZ3bbOpprwyc/z/OtRx2ziNY0woXqJfN0Ak7MfNhaZPaudFBg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:18:45.136755Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.08132","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d9b2e119f8279cb7b1b8efa0b8339655cb29fd48944cf897cd03eaa3aa0a075d","sha256:11cde9e0576681cf493d8278c2111acf7309b797e3e317c5830e9255031b2caf"],"state_sha256":"28b3cbe1b4d817552e94ef70c2fee0f4141c7eba7f2c232af93c792bcd67d5c8"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"PKEqpTi467oHmfy5JYdB6QO//AHzRARxkqGuI6qpNkSdXW1/PvgVMOp0W2x+Q3qVhxPN+PxvJ5yvDX3P6d4NAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-18T06:12:17.448890Z","bundle_sha256":"ae74479c77dc8f65b7b92069bb3078eea82d974676e31ea6699019da81618823"}}