{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:KISKAND25DFFKUMHJFNAHGIKD4","short_pith_number":"pith:KISKAND2","schema_version":"1.0","canonical_sha256":"5224a0347ae8ca555187495a03990a1f2df44a81789eb8f99e04076505e95759","source":{"kind":"arxiv","id":"2607.22242","version":1},"attestation_state":"computed","paper":{"title":"Agentic CPU-GPU Scheduling for Heterogeneous AI Workloads","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.DC","authors_text":"Sherief Reda, Tianxi Lu","submitted_at":"2026-07-24T12:18:36Z","abstract_excerpt":"Agentic AI systems compose heterogeneous tool workloads on shared GPU/CPU infrastructure, yet existing frameworks assign all GPU-capable tools to the GPU by default. We profile 19 AI tools across GPU and CPU and find that 11 are GPU-preferred, 4 are ambiguous, 1 is CPU-preferred due to PCIe transfer dominance, and 3 are device-neutral, establishing that blanket GPU-first scheduling is suboptimal. We formulate device scheduling as assigning each tool to one of three options: immediate GPU execution, queued GPU execution, or CPU offload, under a VRAM budget, and identify two runtime factors that"},"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":"2607.22242","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2026-07-24T12:18:36Z","cross_cats_sorted":[],"title_canon_sha256":"b6160f5d8dd8dee7b198e83448dbd5fda2ad654a14adfdd3de8789382a446a70","abstract_canon_sha256":"5cb4a2e3458f1e552f669a0b952713fc04fe5745fcfe8a34aaf51c4dc38811e4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-27T01:20:55.577230Z","signature_b64":"YH/7WQcQ1TYkzvvNJfESujTYb6ORR2loyjZSJSCfH8ptQPcE2ud4JHIijpZHL1P9sFMFoxoJKVsWTS+7j3PoAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5224a0347ae8ca555187495a03990a1f2df44a81789eb8f99e04076505e95759","last_reissued_at":"2026-07-27T01:20:55.576378Z","signature_status":"signed_v1","first_computed_at":"2026-07-27T01:20:55.576378Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Agentic CPU-GPU Scheduling for Heterogeneous AI Workloads","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.DC","authors_text":"Sherief Reda, Tianxi Lu","submitted_at":"2026-07-24T12:18:36Z","abstract_excerpt":"Agentic AI systems compose heterogeneous tool workloads on shared GPU/CPU infrastructure, yet existing frameworks assign all GPU-capable tools to the GPU by default. We profile 19 AI tools across GPU and CPU and find that 11 are GPU-preferred, 4 are ambiguous, 1 is CPU-preferred due to PCIe transfer dominance, and 3 are device-neutral, establishing that blanket GPU-first scheduling is suboptimal. We formulate device scheduling as assigning each tool to one of three options: immediate GPU execution, queued GPU execution, or CPU offload, under a VRAM budget, and identify two runtime factors that"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.22242","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/2607.22242/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":"2607.22242","created_at":"2026-07-27T01:20:55.576817+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.22242v1","created_at":"2026-07-27T01:20:55.576817+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.22242","created_at":"2026-07-27T01:20:55.576817+00:00"},{"alias_kind":"pith_short_12","alias_value":"KISKAND25DFF","created_at":"2026-07-27T01:20:55.576817+00:00"},{"alias_kind":"pith_short_16","alias_value":"KISKAND25DFFKUMH","created_at":"2026-07-27T01:20:55.576817+00:00"},{"alias_kind":"pith_short_8","alias_value":"KISKAND2","created_at":"2026-07-27T01:20:55.576817+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/KISKAND25DFFKUMHJFNAHGIKD4","json":"https://pith.science/pith/KISKAND25DFFKUMHJFNAHGIKD4.json","graph_json":"https://pith.science/api/pith-number/KISKAND25DFFKUMHJFNAHGIKD4/graph.json","events_json":"https://pith.science/api/pith-number/KISKAND25DFFKUMHJFNAHGIKD4/events.json","paper":"https://pith.science/paper/KISKAND2"},"agent_actions":{"view_html":"https://pith.science/pith/KISKAND25DFFKUMHJFNAHGIKD4","download_json":"https://pith.science/pith/KISKAND25DFFKUMHJFNAHGIKD4.json","view_paper":"https://pith.science/paper/KISKAND2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.22242&json=true","fetch_graph":"https://pith.science/api/pith-number/KISKAND25DFFKUMHJFNAHGIKD4/graph.json","fetch_events":"https://pith.science/api/pith-number/KISKAND25DFFKUMHJFNAHGIKD4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KISKAND25DFFKUMHJFNAHGIKD4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KISKAND25DFFKUMHJFNAHGIKD4/action/storage_attestation","attest_author":"https://pith.science/pith/KISKAND25DFFKUMHJFNAHGIKD4/action/author_attestation","sign_citation":"https://pith.science/pith/KISKAND25DFFKUMHJFNAHGIKD4/action/citation_signature","submit_replication":"https://pith.science/pith/KISKAND25DFFKUMHJFNAHGIKD4/action/replication_record"}},"created_at":"2026-07-27T01:20:55.576817+00:00","updated_at":"2026-07-27T01:20:55.576817+00:00"}