Skip to content
AI Atlas
Papercs.AI

MAPS: Memory-Aware Predictive Scheduling Framework for Large Language Model Serving

Published 15 Sept 2026arXiv:2609.15359

data quality89

Updated 29 h ago · first seen 15 Sept 2026

paper_01M2JK195CPWM5P7A93GCJT5BS

Abstract

The surge of large language model (LLM) applications on personal devices imposes massive, bursty workloads on cloud serving infrastructure. While prefill-decode disaggregation improves throughput and scalability, memory-bound decode instances often suffer from persistent load imbalance, as output lengths are unknown when requests arrive at the cloud. To address this, we propose MAPS, a Memory-Aware Predictive Scheduling framework tailored for disaggregated LLM serving. MAPS performs device-assisted speculative output length prediction overlapped with cloud-side prefilling, incurring negligible latency overhead. To handle generation uncertainty, MAPS applies uncertainty-aware calibration to derive output-length upper bounds with target coverage, enabling safe scheduling decisions. Building on these bounds, MAPS employs a hierarchical global-local scheduling strategy to mitigate inter-decoder queue buildup and intra-decoder head-of-line blocking. Extensive experiments on two real-world workloads and two LLMs show that MAPS significantly outperforms three state-of-the-art systems, reducing average end-to-end latency by 42.6 and tail latency by up to 84.8.

Authors

Authors 6

Cheng ZhangShaoyuan HuangTiancheng ZhangXiaofei WangYulin ChenYunfeng Zhao

Linked names open researcher pages (created from the paper's author list; name-only, no affiliation unless a source states it). Unlinked names have no researcher record yet.

Organizations

Organizations 0

No organization stated. arXiv metadata does not carry affiliations; an organization is linked only when a model card or lab page cites the paper.

Models

Models introduced or described 0

Inbound described_by relations from model cards and documentation.

No model links this paper yet

Model pages link papers through their model cards and documentation; the relation is written only when a source states it.

Datasets

Datasets used 0

No dataset relation recorded.

Benchmarks

Benchmarks used 0

No benchmark relation recorded.

Code

Repositories & frameworks 0

No repository linked.

Timeline

Timeline 1

Full timeline →

Sources

Sources 1

Source documents
SourceDocumentTypeTierLast observedSnapshots
arXiv (Atom API + RSS)rss.arxiv.org/rss/cs.AI feedT1· Official21 h ago4

Tier 1 = official/primary, 2 = quality secondary, 3 = community, 4 = unverified. Every snapshot is archived; see all sources and the methodology.