Language models help but don’t fully control edge cloud service orchestration
Can We Rely on Language Models to Orchestrate Services across the Edge-Cloud Continuum?
Distributed, Parallel, and Cluster Computing
Summary
Managing services that span both local devices (the edge) and distant cloud servers is complicated. The authors reviewed 39 studies to see how language models (like AI that understands human language) are being used to help with decisions in this setup. They found these models often assist by interpreting information and suggesting plans, but other parts of the system usually make final decisions and enforce rules. The authors conclude that trusting language models alone isn’t enough without careful controls and evidence that the system behaves well.
What this means in practice
- •For cloud service operators: Design orchestration workflows that use language models for planning but keep strict controls on execution authority to ensure reliability.
- •For network infrastructure engineers: Integrate language model outputs with system safeguards when managing heterogeneous edge-cloud environments to improve diagnosis and remediation.
A survey. It maps existing work.
Authors
Zhenghao Wu, Ruihai Dong, Reza Farahani, Hamed Haddadi, Omer Rana, Hadi Tabatabaee Malazi
Abstract
Service orchestration across the edge--cloud continuum increasingly uses language models (LMs) to translate intents and operational evidence into scheduling, configuration, diagnosis, and remediation decisions. Yet, useful model outputs do not by themselves establish dependable orchestration: reliability also depends on where the LM enters the decision process, what authority it is granted, how actions are constrained, and what system-level evidence supports the claimed effect. Existing surveys provide limited synthesis of the links between continuum-specific orchestration decisions, LM responsibility, operational authority, assurance controls, and supporting empirical evidence. We systematically review 39 primary studies and develop a five-dimensional analytical framework for connecting system context, orchestration decisions, LM agency, operational safeguards, and evaluation evidence. The synthesis reveals predominantly hybrid designs in which LMs support interpretation, candidate generation, and mitigation planning, while downstream components commonly retain feasibility enforcement and execution authority. Bounded closed-loop remediation is emerging, but continuum-specific constraints remain unevenly represented in decision formulations, and production-scale evidence across heterogeneous edge--cloud environments remains limited. These findings suggest that reliance on LM-enabled orchestration depends on aligning decision constraints, operational authority, safeguards, and evaluation evidence with the claimed system effect.