Task-Centric IoT Service Orchestration Across the Edge–Fog–Cloud Continuum: A Structured Review and 6G-Oriented Framework

Abstract

The rapid growth of Internet of Things (IoT) deployments is producing workloads whose latency, bandwidth, energy, privacy, and reliability requirements vary over time. Cloud-centric processing provides substantial compute and storage capacity, but device-to-cloud communication can increase delay and network traffic for time-sensitive services. Edge, fog, and multi-access edge computing (MEC) reduce these limitations by moving selected processing closer to data sources, while the edge-fog-cloud continuum enables placement across heterogeneous resources. This paper presents a structured review of IoT service orchestration across the continuum, with emphasis on task placement and the factors used in orchestration decisions. The review is organized around five research questions covering architectural evolution, orchestration approaches, decision inputs, evaluation practices, and 6G requirements. A task-centric synthesis shows how urgency, workload characteristics, network state, resource availability, and other
Constraints influence placement. Based on these findings, a reference framework is presented that combines a rulebased priority engine for urgent events with a machine-learning decision layer for non-urgent workloads. The review also identifies open challenges in explainability, data availability, multi-objective optimization, interoperability, security, and 6G-aware orchestration, and it defines an evaluation methodology for future prototype validation.