{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:U6A4KXFOFRNGAENONH3ZZKI2TX","short_pith_number":"pith:U6A4KXFO","schema_version":"1.0","canonical_sha256":"a781c55cae2c5a6011ae69f79ca91a9dc5cf27f9c8bfacd23e810154eee1e58e","source":{"kind":"arxiv","id":"2606.00735","version":1},"attestation_state":"computed","paper":{"title":"ViBE: Co-Optimizing Workload Skew and Hardware Variability for MoE Serving","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.DC","authors_text":"Divya Mahajan, Ephrem Wu, Marko Scrbak, Seokjin Go, Srilatha Manne","submitted_at":"2026-05-30T13:57:09Z","abstract_excerpt":"In distributed Mixture-of-Experts (MoE) inference, input-dependent token routing interacts with GPU performance variability to create persistent stragglers under synchronized execution, where the slowest GPU determines layer latency. This performance variability is inherent to modern accelerators: manufacturing variation, power limits, and thermal conditions introduce measurable execution-time differences across nominally identical GPUs. The core challenge is that MoE execution-time imbalance arises from the interaction of workload skew and hardware asymmetry. Token routing produces uneven and"},"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":"2606.00735","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2026-05-30T13:57:09Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"d4cff576410ab0022fc9c8d58964d25f2bc4da21bdfe314c0760f967f2f41a36","abstract_canon_sha256":"12a2697d3af9cf7688cb5c02319340900f68f12498fc75d01ea463c6ba3bee6a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T01:04:04.179184Z","signature_b64":"flMQI/COKLL1UETauygpXlxi7cZXyyYhJyP2G40ryO9wtaefpzdZzy+6uzRlNbdtwtUt5GRgwdbwRoL3LvnuCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a781c55cae2c5a6011ae69f79ca91a9dc5cf27f9c8bfacd23e810154eee1e58e","last_reissued_at":"2026-06-02T01:04:04.178759Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T01:04:04.178759Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ViBE: Co-Optimizing Workload Skew and Hardware Variability for MoE Serving","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.DC","authors_text":"Divya Mahajan, Ephrem Wu, Marko Scrbak, Seokjin Go, Srilatha Manne","submitted_at":"2026-05-30T13:57:09Z","abstract_excerpt":"In distributed Mixture-of-Experts (MoE) inference, input-dependent token routing interacts with GPU performance variability to create persistent stragglers under synchronized execution, where the slowest GPU determines layer latency. This performance variability is inherent to modern accelerators: manufacturing variation, power limits, and thermal conditions introduce measurable execution-time differences across nominally identical GPUs. The core challenge is that MoE execution-time imbalance arises from the interaction of workload skew and hardware asymmetry. Token routing produces uneven and"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.00735","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/2606.00735/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":"2606.00735","created_at":"2026-06-02T01:04:04.178833+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.00735v1","created_at":"2026-06-02T01:04:04.178833+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.00735","created_at":"2026-06-02T01:04:04.178833+00:00"},{"alias_kind":"pith_short_12","alias_value":"U6A4KXFOFRNG","created_at":"2026-06-02T01:04:04.178833+00:00"},{"alias_kind":"pith_short_16","alias_value":"U6A4KXFOFRNGAENO","created_at":"2026-06-02T01:04:04.178833+00:00"},{"alias_kind":"pith_short_8","alias_value":"U6A4KXFO","created_at":"2026-06-02T01:04:04.178833+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/U6A4KXFOFRNGAENONH3ZZKI2TX","json":"https://pith.science/pith/U6A4KXFOFRNGAENONH3ZZKI2TX.json","graph_json":"https://pith.science/api/pith-number/U6A4KXFOFRNGAENONH3ZZKI2TX/graph.json","events_json":"https://pith.science/api/pith-number/U6A4KXFOFRNGAENONH3ZZKI2TX/events.json","paper":"https://pith.science/paper/U6A4KXFO"},"agent_actions":{"view_html":"https://pith.science/pith/U6A4KXFOFRNGAENONH3ZZKI2TX","download_json":"https://pith.science/pith/U6A4KXFOFRNGAENONH3ZZKI2TX.json","view_paper":"https://pith.science/paper/U6A4KXFO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.00735&json=true","fetch_graph":"https://pith.science/api/pith-number/U6A4KXFOFRNGAENONH3ZZKI2TX/graph.json","fetch_events":"https://pith.science/api/pith-number/U6A4KXFOFRNGAENONH3ZZKI2TX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/U6A4KXFOFRNGAENONH3ZZKI2TX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/U6A4KXFOFRNGAENONH3ZZKI2TX/action/storage_attestation","attest_author":"https://pith.science/pith/U6A4KXFOFRNGAENONH3ZZKI2TX/action/author_attestation","sign_citation":"https://pith.science/pith/U6A4KXFOFRNGAENONH3ZZKI2TX/action/citation_signature","submit_replication":"https://pith.science/pith/U6A4KXFOFRNGAENONH3ZZKI2TX/action/replication_record"}},"created_at":"2026-06-02T01:04:04.178833+00:00","updated_at":"2026-06-02T01:04:04.178833+00:00"}