{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,5]],"date-time":"2026-08-05T06:45:20Z","timestamp":1785912320521,"version":"3.56.0"},"reference-count":51,"publisher":"IEEE","license":[{"start":{"date-parts":[[2026,6,1]],"date-time":"2026-06-01T00:00:00Z","timestamp":1780272000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2026,6,1]],"date-time":"2026-06-01T00:00:00Z","timestamp":1780272000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,6,1]],"date-time":"2026-06-01T00:00:00Z","timestamp":1780272000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"name":"National Key R&D Program of China","award":["2024YFB4505604"],"award-info":[{"award-number":["2024YFB4505604"]}]},{"name":"National Natural Science Foundation of China","award":["62402024"],"award-info":[{"award-number":["62402024"]}]},{"name":"Fundamental Research Funds for the Central Universities"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,6]]},"DOI":"10.1109\/2575-8411.2026.00031","type":"proceedings-article","created":{"date-parts":[[2026,8,4]],"date-time":"2026-08-04T19:16:15Z","timestamp":1785870975000},"page":"250-260","source":"Crossref","is-referenced-by-count":0,"title":["Maestro: Workload-Aware Cross-Cluster Scheduling for LLM-based Multi-Agent Systems"],"prefix":"10.1109","author":[{"given":"Jinghao","family":"Wang","sequence":"first","affiliation":[{"name":"Beihang University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiao","family":"Zhou","sequence":"additional","affiliation":[{"name":"Beihang University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiaoyang","family":"Sun","sequence":"additional","affiliation":[{"name":"University of Leeds"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yihui","family":"Zhang","sequence":"additional","affiliation":[{"name":"Beihang University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yilong","family":"Li","sequence":"additional","affiliation":[{"name":"Beihang University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tianyu","family":"Wo","sequence":"additional","affiliation":[{"name":"Beihang University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xu","family":"Wang","sequence":"additional","affiliation":[{"name":"Beihang University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chunming","family":"Hu","sequence":"additional","affiliation":[{"name":"Beihang University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Renyu","family":"Yang","sequence":"additional","affiliation":[{"name":"Beihang University"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1016\/j.peva.2024.102451"},{"key":"ref2","article-title":"Internet Access Performance","journal-title":"Alibaba Cloud documentation"},{"key":"ref3","article-title":"Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback","author":"Bai","year":"2022","journal-title":"arXiv"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2025.acl-long.183"},{"key":"ref5","article-title":"AgentVerse: Facilitating MultiAgent Collaboration and Exploring Emergent Behaviors","author":"Chen","year":"2024","journal-title":"ICLR"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/ICWS62655.2024.00104"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM55648.2025.11044657"},{"key":"ref8","article-title":"Improving factuality and reasoning in language models through multiagent debate","author":"Du","year":"2024","journal-title":"ICML"},{"key":"ref9","article-title":"MuxServe: flexible spatialtemporal multiplexing for multiple LLM serving","author":"Duan","year":"2024","journal-title":"ICML"},{"key":"ref10","article-title":"Serving DNNs like Clockwork: Performance Predictability from the Bottom Up","author":"Gujarati","year":"2020","journal-title":"USENIX OSDI"},{"key":"ref11","article-title":"DeepSeek-R1: Incentivizing reasoning capability in LLMs via reinforcement learning","author":"Guo","year":"2025","journal-title":"arXiv"},{"key":"ref12","article-title":"Large language model based multiagents: a survey of progress and challenges","author":"Guo","year":"2024","journal-title":"IJCAI"},{"key":"ref13","article-title":"Measuring Mathematical Problem Solving With the MATH Dataset","author":"Hendrycks","year":"2021","journal-title":"NeurIPS"},{"key":"ref14","article-title":"MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework","author":"Hong","year":"2024","journal-title":"ICLR"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.52202\/075280-0791"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1145\/3600006.3613165"},{"key":"ref17","article-title":"Parrot: Efficient serving of LLMbased applications with semantic variable","author":"Lin","year":"2024","journal-title":"OSDI"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1145\/3803390"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1145\/3698038.3698532"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/TCC.2025.3624752"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM55648.2025.11044761"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1145\/3689031.3717459"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/TMC.2025.3527707"},{"key":"ref24","article-title":"Efficient MoE Inference with FineGrained Scheduling of Disaggregated Expert Parallelism","author":"Pan","year":"2025","journal-title":"arXiv"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1145\/3698038.3698523"},{"key":"ref26","article-title":"Efficient Interactive LLM Serving with Proxy Model-based Sequence Length Prediction","volume-title":"ASPLOS Cloud Intelligence\/AIOps Workshop","author":"Qiu","year":"2024"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D16-1264"},{"key":"ref28","article-title":"GPQA: A Graduate-Level GoogleProof Q&A Benchmark","author":"Rein","year":"2023","journal-title":"arXiv"},{"key":"ref29","article-title":"FlexGen: High-Throughput Generative Inference of Large Language Models with a Single GPU","author":"Sheng","year":"2023","journal-title":"ICML"},{"key":"ref30","article-title":"IBM HR Analytics Employee Attrition & Performance","author":"Subhash","year":"2017","journal-title":"Kaggle dataset"},{"key":"ref31","article-title":"AIBrix: Towards Scalable, CostEffective Large Language Model Inference Infrastructure","year":"2025","journal-title":"arXiv"},{"key":"ref32","article-title":"LLaMA: Open and Efficient Foundation Language Models","author":"Touvron","year":"2023","journal-title":"arXiv"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1145\/3749483"},{"key":"ref34","article-title":"Mixture-of-agents enhances large language model capabilities","author":"Wang","year":"2024","journal-title":"arXiv"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.acl-long.147"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1109\/ASE63991.2025.00263"},{"key":"ref37","article-title":"Fast Distributed Inference Serving for Large Language Models","author":"Wu","year":"2024","journal-title":"arXiv"},{"key":"ref38","article-title":"AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation","author":"Wu","year":"2023","journal-title":"arXiv"},{"key":"ref39","article-title":"Qwen3 Technical Report","author":"Yang","year":"2025","journal-title":"arXiv"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/ASE63991.2025.00311"},{"key":"ref41","article-title":"ReAct: Synergizing reasoning and acting in language models","author":"Yao","year":"2023","journal-title":"ICLR"},{"key":"ref42","article-title":"Orca: A Distributed Serving System for Transformer-Based Generative Models","author":"Yu","year":"2022","journal-title":"USENIX OSDI"},{"key":"ref43","article-title":"Torpor: GPU-enabled serverless computing for low-latency, resource-efficient inference","author":"Yu","year":"2025","journal-title":"USENIX ATC"},{"key":"ref44","volume-title":"Prism: Unleashing GPU Sharing for Cost-Efficient Multi-LLM Serving","author":"Yu","year":"2026"},{"key":"ref45","article-title":"OpenCSG Chinese Corpus: A Series of High-quality Chinese Datasets for LLM Training","author":"Yu","year":"2025","journal-title":"arXiv"},{"key":"ref46","article-title":"Cauchy: A Cost-Efficient LLM Serving System through Adaptive Heterogeneous Deployment","author":"Zhang","year":"2025","journal-title":"ACM SoCC"},{"key":"ref47","article-title":"Cuckoo: Deadline-Aware Job Packing on Heterogeneous GPUs for DL Model Training","author":"Zhang","year":"2025","journal-title":"ACM SoCC"},{"key":"ref48","article-title":"Enabling Disaggregated MultiStage MLLM Inference via GPU-Internal Scheduling and Resource Sharing","author":"Zhao","year":"2025","journal-title":"arXiv"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.52202\/075280-2020"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.52202\/079017-2000"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM55648.2025.11044693"}],"event":{"name":"2026 IEEE 46th International Conference on Distributed Computing Systems (ICDCS)","location":"Seoul, Korea, Republic of","start":{"date-parts":[[2026,6,22]]},"end":{"date-parts":[[2026,6,25]]}},"container-title":["2026 IEEE 46th International Conference on Distributed Computing Systems (ICDCS)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11618978\/11618981\/11619200.pdf?arnumber=11619200","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,5]],"date-time":"2026-08-05T05:53:16Z","timestamp":1785909196000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11619200\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6]]},"references-count":51,"URL":"https:\/\/doi.org\/10.1109\/2575-8411.2026.00031","relation":{},"subject":[],"published":{"date-parts":[[2026,6]]}}}