{"paper":{"title":"Sustainable Graph Analytics Workload Scheduling with Evolutionary Reinforcement Learning in Edge-Cloud Systems","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"A hybrid evolutionary reinforcement learning scheduler reduces SLA violations by up to 45 percent and carbon emissions by up to 12 percent for graph analytics in edge-cloud systems.","cross_cats":[],"primary_cat":"cs.DC","authors_text":"A. Islam, C. Bash, D. Milojicic, H. Moore, M. Ghose, P. Ramicetty, S. Pasricha, S. Qi","submitted_at":"2026-05-13T13:15:11Z","abstract_excerpt":"Graph analytics powers modern intelligent systems such as smart cities, cyber-physical infrastructure, IoT security, and large-scale social networks. As these workloads scale in complexity, their execution in heterogeneous edge-cloud environments results in higher energy use and carbon emission footprint. To address this challenge, we propose MERSEM, a multi-objective evolutionary reinforcement learning framework for sustainable edge-cloud system management. MERSEM integrates evolutionary search with reinforcement learning (RL) to solve the problem of graph workload allocation and scheduling. "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Experimental results demonstrate that MERSEM outperforms the state-of-the-art with up to 45% SLA violation reductions and up to 12% carbon emission reductions.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The simulation environment used for experiments accurately captures real-world dynamics of carbon intensity, resource heterogeneity, and workload arrival patterns in edge-cloud systems.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"MERSEM uses evolutionary reinforcement learning to allocate graph workloads in edge-cloud systems, reducing SLA violations by up to 45% and carbon emissions by up to 12% versus prior methods.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A hybrid evolutionary reinforcement learning scheduler reduces SLA violations by up to 45 percent and carbon emissions by up to 12 percent for graph analytics in edge-cloud systems.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"f47caa8a0b85eadd441062d129577075e474c6e9144171e41f90ec5e22e60272"},"source":{"id":"2605.13489","kind":"arxiv","version":1},"verdict":{"id":"3c02257c-9a75-44cd-acdd-88774e1b259c","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T19:25:30.193358Z","strongest_claim":"Experimental results demonstrate that MERSEM outperforms the state-of-the-art with up to 45% SLA violation reductions and up to 12% carbon emission reductions.","one_line_summary":"MERSEM uses evolutionary reinforcement learning to allocate graph workloads in edge-cloud systems, reducing SLA violations by up to 45% and carbon emissions by up to 12% versus prior methods.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The simulation environment used for experiments accurately captures real-world dynamics of carbon intensity, resource heterogeneity, and workload arrival patterns in edge-cloud systems.","pith_extraction_headline":"A hybrid evolutionary reinforcement learning scheduler reduces SLA violations by up to 45 percent and carbon emissions by up to 12 percent for graph analytics in edge-cloud systems."},"references":{"count":18,"sample":[{"doi":"10.1016/j.egyr","year":2025,"title":"Alsharif et al","work_id":"a64ee313-1915-465d-9a71-c9d2287cafb1","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.1145/3773274.3774279","year":2026,"title":"In: Proceedings of the 18th IEEE/ACM International Conference on Utility and Cloud Com- puting","work_id":"1f743599-258c-417c-a85c-8ea77ccf927e","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2002,"title":"A fast and elitist multiobjective genetic algorithm: NSGA-II","work_id":"5811a599-5539-4635-a89b-72aac96c0052","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.1109/tevc.2013.2281535","year":2014,"title":"Montes de Oca, Thomas Stützle, and Marco Dorigo","work_id":"4d641aed-fd6d-4395-a645-5732fa4e0407","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.1007/s10586-021-03307-2","year":2021,"title":"Etemadi et al","work_id":"b3dba67a-1a66-457d-80d5-94ce473a1296","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":18,"snapshot_sha256":"f5866003db7b5303eb901e45638448270f45abc77d2dc3895951eabd54b80b52","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"}