{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:R6GACQN7Z5OB3VUCJ3UPDSCPXE","short_pith_number":"pith:R6GACQN7","schema_version":"1.0","canonical_sha256":"8f8c0141bfcf5c1dd6824ee8f1c84fb931c50e0e6e39de51b717831ad46327b9","source":{"kind":"arxiv","id":"2607.26633","version":1},"attestation_state":"computed","paper":{"title":"NELSSA: A GPU-PNM Heterogeneous System for Mixed-Length LLM Serving via Length-based Request Placement","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AR","authors_text":"Donguk Moon, Euiseok Kim, Guseul Heo, Hoshik Kim, Hyeongseok Gwak, Jongse Park, Junseok Lee, Kangkyu Park, Kwangsik Shin, Myunghyun Rhee, Seungyong Lee, Sookyung Choi, Youngpyo Joo, Yunseo Chun","submitted_at":"2026-07-29T08:58:04Z","abstract_excerpt":"Modern LLMs and their agentic applications are broadening the range of serving workloads, spanning context lengths from a few hundred tokens to hundreds of thousands. As these requests frequently interleave within the same serving window, LLM serving systems must handle highly heterogeneous mixed-length workloads. Such mixed-length workloads expose fundamental inefficiencies in GPU-centric serving architectures, whose throughput depends on large, memory-constrained batches. In this paper, we present NELSSA, an LLM serving system that integrates GPUs with real-world Processing-near-Memory (PNM)"},"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":"2607.26633","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AR","submitted_at":"2026-07-29T08:58:04Z","cross_cats_sorted":[],"title_canon_sha256":"0a93a664eca161a8398332b68f6510332976170ad3ed037df33e364b10c27c11","abstract_canon_sha256":"54ca0fe4ce2caa61a65060b4168ab158bd5174b7e33e762d87bd51d7b1fab178"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8f8c0141bfcf5c1dd6824ee8f1c84fb931c50e0e6e39de51b717831ad46327b9","last_reissued_at":"2026-07-30T01:21:38.317044Z","signature_status":"unsigned_v0","first_computed_at":"2026-07-30T01:21:38.317044Z"},"graph_snapshot":{"paper":{"title":"NELSSA: A GPU-PNM Heterogeneous System for Mixed-Length LLM Serving via Length-based Request Placement","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AR","authors_text":"Donguk Moon, Euiseok Kim, Guseul Heo, Hoshik Kim, Hyeongseok Gwak, Jongse Park, Junseok Lee, Kangkyu Park, Kwangsik Shin, Myunghyun Rhee, Seungyong Lee, Sookyung Choi, Youngpyo Joo, Yunseo Chun","submitted_at":"2026-07-29T08:58:04Z","abstract_excerpt":"Modern LLMs and their agentic applications are broadening the range of serving workloads, spanning context lengths from a few hundred tokens to hundreds of thousands. As these requests frequently interleave within the same serving window, LLM serving systems must handle highly heterogeneous mixed-length workloads. Such mixed-length workloads expose fundamental inefficiencies in GPU-centric serving architectures, whose throughput depends on large, memory-constrained batches. In this paper, we present NELSSA, an LLM serving system that integrates GPUs with real-world Processing-near-Memory (PNM)"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.26633","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/2607.26633/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":"2607.26633","created_at":"2026-07-30T01:21:38.322165+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.26633v1","created_at":"2026-07-30T01:21:38.322165+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.26633","created_at":"2026-07-30T01:21:38.322165+00:00"},{"alias_kind":"pith_short_12","alias_value":"R6GACQN7Z5OB","created_at":"2026-07-30T01:21:38.322165+00:00"},{"alias_kind":"pith_short_16","alias_value":"R6GACQN7Z5OB3VUC","created_at":"2026-07-30T01:21:38.322165+00:00"},{"alias_kind":"pith_short_8","alias_value":"R6GACQN7","created_at":"2026-07-30T01:21:38.322165+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/R6GACQN7Z5OB3VUCJ3UPDSCPXE","json":"https://pith.science/pith/R6GACQN7Z5OB3VUCJ3UPDSCPXE.json","graph_json":"https://pith.science/api/pith-number/R6GACQN7Z5OB3VUCJ3UPDSCPXE/graph.json","events_json":"https://pith.science/api/pith-number/R6GACQN7Z5OB3VUCJ3UPDSCPXE/events.json","paper":"https://pith.science/paper/R6GACQN7"},"agent_actions":{"view_html":"https://pith.science/pith/R6GACQN7Z5OB3VUCJ3UPDSCPXE","download_json":"https://pith.science/pith/R6GACQN7Z5OB3VUCJ3UPDSCPXE.json","view_paper":"https://pith.science/paper/R6GACQN7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.26633&json=true","fetch_graph":"https://pith.science/api/pith-number/R6GACQN7Z5OB3VUCJ3UPDSCPXE/graph.json","fetch_events":"https://pith.science/api/pith-number/R6GACQN7Z5OB3VUCJ3UPDSCPXE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/R6GACQN7Z5OB3VUCJ3UPDSCPXE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/R6GACQN7Z5OB3VUCJ3UPDSCPXE/action/storage_attestation","attest_author":"https://pith.science/pith/R6GACQN7Z5OB3VUCJ3UPDSCPXE/action/author_attestation","sign_citation":"https://pith.science/pith/R6GACQN7Z5OB3VUCJ3UPDSCPXE/action/citation_signature","submit_replication":"https://pith.science/pith/R6GACQN7Z5OB3VUCJ3UPDSCPXE/action/replication_record"}},"created_at":"2026-07-30T01:21:38.322165+00:00","updated_at":"2026-07-30T01:21:38.322165+00:00"}