{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:L5U5XIW7WFOYQBD75Z3IK3G77K","short_pith_number":"pith:L5U5XIW7","schema_version":"1.0","canonical_sha256":"5f69dba2dfb15d88047fee76856cdffab04c441a7306175c091d7d37ba8dcd35","source":{"kind":"arxiv","id":"2412.12487","version":1},"attestation_state":"computed","paper":{"title":"Echo: Simulating Distributed Training At Scale","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC"],"primary_cat":"cs.LG","authors_text":"Chuan Wu, Hong Xu, Jingzong Li, Kaiwen Chen, Peng Cheng, Tianyuan Wu, Tsung-Yi Ho, Wei Wang, Yicheng Feng, Yuetao Chen","submitted_at":"2024-12-17T02:44:35Z","abstract_excerpt":"Simulation offers unique values for both enumeration and extrapolation purposes, and is becoming increasingly important for managing the massive machine learning (ML) clusters and large-scale distributed training jobs. In this paper, we build Echo to tackle three key challenges in large-scale training simulation: (1) tracing the runtime training workloads at each device in an ex-situ fashion so we can use a single device to obtain the actual execution graphs of 1K-GPU training, (2) accurately estimating the collective communication without high overheads of discrete-event based network simulat"},"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":"2412.12487","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-12-17T02:44:35Z","cross_cats_sorted":["cs.DC"],"title_canon_sha256":"e2d44b9f9d6d60f62a4ded41a8600af20323cf50937810dab41a8f5f1a36f790","abstract_canon_sha256":"c73d7c29c6c386faf26b547e50fad7b057393bdb5a85bdbd9dd65655abd49d9b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:50:15.667442Z","signature_b64":"TtxEkDojjfzmz4/3PH+PPYlDTJ1k3KCLOy+L0bcXVOvIGnJ79ok+XCH4gK+vGD1nA3d/ZlaV0NKi0KClxACZDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5f69dba2dfb15d88047fee76856cdffab04c441a7306175c091d7d37ba8dcd35","last_reissued_at":"2026-07-05T09:50:15.666969Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:50:15.666969Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Echo: Simulating Distributed Training At Scale","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC"],"primary_cat":"cs.LG","authors_text":"Chuan Wu, Hong Xu, Jingzong Li, Kaiwen Chen, Peng Cheng, Tianyuan Wu, Tsung-Yi Ho, Wei Wang, Yicheng Feng, Yuetao Chen","submitted_at":"2024-12-17T02:44:35Z","abstract_excerpt":"Simulation offers unique values for both enumeration and extrapolation purposes, and is becoming increasingly important for managing the massive machine learning (ML) clusters and large-scale distributed training jobs. In this paper, we build Echo to tackle three key challenges in large-scale training simulation: (1) tracing the runtime training workloads at each device in an ex-situ fashion so we can use a single device to obtain the actual execution graphs of 1K-GPU training, (2) accurately estimating the collective communication without high overheads of discrete-event based network simulat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.12487","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/2412.12487/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":"2412.12487","created_at":"2026-07-05T09:50:15.667026+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.12487v1","created_at":"2026-07-05T09:50:15.667026+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.12487","created_at":"2026-07-05T09:50:15.667026+00:00"},{"alias_kind":"pith_short_12","alias_value":"L5U5XIW7WFOY","created_at":"2026-07-05T09:50:15.667026+00:00"},{"alias_kind":"pith_short_16","alias_value":"L5U5XIW7WFOYQBD7","created_at":"2026-07-05T09:50:15.667026+00:00"},{"alias_kind":"pith_short_8","alias_value":"L5U5XIW7","created_at":"2026-07-05T09:50:15.667026+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.26633","citing_title":"Simulating Unified Tensor Resharding in heterogeneous AI systems","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2606.03077","citing_title":"Libra: Efficient Resource Management for Agentic RL Post-Training","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17164","citing_title":"Charon: A Unified and Fine-Grained Simulator for Large-Scale LLM Training and Inference","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2605.21312","citing_title":"Frontier: Towards Comprehensive and Accurate LLM Inference Simulation","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17164","citing_title":"Charon: A Unified and Fine-Grained Simulator for Large-Scale LLM Training and Inference","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2604.12090","citing_title":"Evaluating Cross-Architecture Performance Modeling of Distributed ML Workloads Using StableHLO","ref_index":30,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/L5U5XIW7WFOYQBD75Z3IK3G77K","json":"https://pith.science/pith/L5U5XIW7WFOYQBD75Z3IK3G77K.json","graph_json":"https://pith.science/api/pith-number/L5U5XIW7WFOYQBD75Z3IK3G77K/graph.json","events_json":"https://pith.science/api/pith-number/L5U5XIW7WFOYQBD75Z3IK3G77K/events.json","paper":"https://pith.science/paper/L5U5XIW7"},"agent_actions":{"view_html":"https://pith.science/pith/L5U5XIW7WFOYQBD75Z3IK3G77K","download_json":"https://pith.science/pith/L5U5XIW7WFOYQBD75Z3IK3G77K.json","view_paper":"https://pith.science/paper/L5U5XIW7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.12487&json=true","fetch_graph":"https://pith.science/api/pith-number/L5U5XIW7WFOYQBD75Z3IK3G77K/graph.json","fetch_events":"https://pith.science/api/pith-number/L5U5XIW7WFOYQBD75Z3IK3G77K/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/L5U5XIW7WFOYQBD75Z3IK3G77K/action/timestamp_anchor","attest_storage":"https://pith.science/pith/L5U5XIW7WFOYQBD75Z3IK3G77K/action/storage_attestation","attest_author":"https://pith.science/pith/L5U5XIW7WFOYQBD75Z3IK3G77K/action/author_attestation","sign_citation":"https://pith.science/pith/L5U5XIW7WFOYQBD75Z3IK3G77K/action/citation_signature","submit_replication":"https://pith.science/pith/L5U5XIW7WFOYQBD75Z3IK3G77K/action/replication_record"}},"created_at":"2026-07-05T09:50:15.667026+00:00","updated_at":"2026-07-05T09:50:15.667026+00:00"}