Skip to content
Open access

Mobility-aware task scheduling based on artificial intelligence in edge-cloud systems

Jul 2026 · PeerJ Computer Science · 0 citations · 50 references

TL;DR

A novel hybrid framework that combines a Graph Attention Network (GAT)-based fault prediction model with a Generative Adversarial Network (GAN)-driven task migration decision model is proposed, achieving substantial reductions in task execution time and energy consumption while improving predictive accuracy and resource efficiency.

Abstract

In edge-cloud computing environments, efficient task scheduling and proactive fault management are essential for achieving high performance and resource utilization. This article proposes a novel hybrid framework that combines a Graph Attention Network (GAT)-based fault prediction model with a Generative Adversarial Network (GAN)-driven task migration decision model. The GAT component leverages historical execution data and inter-node dependencies to accurately identify potential failure points, while the GAN component generates optimal migration strategies to preemptively mitigate predicted faults. By integrating proactive fault prediction with intelligent migration decisions, the proposed approach significantly enhances fault tolerance, minimizes service latency, and improves overall system throughput. Extensive experiments conducted on real-world edge-cloud traces demonstrate that our method outperforms state-of-the-art fault prediction and task migration techniques, achieving substantial reductions in task execution time and energy consumption while improving predictive accuracy and resource efficiency.

Read PDF

Similar papers

Open access 2024

AI-Enabled Workload Prediction for Elastic Cloud Computing Systems

In conclusion, the proposed AI-based workload prediction framework significantly enhances cloud resource management by accurately forecasting future workload demands and enabling proactive resource allocation. By utilizing machine learning and deep learning techniques such as LSTM, Random Forest Regression, and Gradient Boosting, the system improves resource utilization, reduces response time, lowers operational costs, and minimizes SLA violations. The results demonstrate superior prediction accuracy compared to traditional methods, leading to better Quality of Service (QoS) and energy efficiency. This study highlights the potential of AI-driven predictive analytics to transform cloud computing from reactive resource management to intelligent, autonomous, and adaptive cloud ecosystems. Future research can further improve performance through the integration of federated learning, reinforcement learning, and edge-cloud computing technologies.

John Peterson · 0 citations
Open access 2024

AI-Assisted Resource Scheduling in Multi-Cloud Computing Environments

This research proposes an AI-driven resource scheduling framework that integrates workload prediction, resource classification, intelligent scheduling, and continuous feedback mechanisms that aims to optimize multiple objectives, including cost reduction, execution efficiency, energy consumption, and SLA compliance.

Michael Anderson · 0 citations
Open access 2025

Context-Aware AI Models for Dynamic Resource Management in Cloud Systems

This study proposes a Context-Aware AI framework for dynamic cloud resource management that incorporates workload patterns, user behavior, network conditions, infrastructure health, and business objectives and provides a foundation for future technologies such as edge computing, IoT, 6G networks, and intelligent enterprise applications.

Richard Evans, Karen Lewis · 0 citations
Open access 2025

Self-Adaptive Distributed Computing Models for High-Performance Analytics

This work proposes a scalable, intelligent, and resilient foundation for next-generation high-performance analytics and data-intensive applications that integrates adaptive resource management, intelligent workload scheduling, dynamic task migration, predictive analytics, and machine learning-based optimization to improve computational efficiency and responsiveness.

John Peterson, L. Martínez · 0 citations
Open access Aug 2026

Hierarchical Scheduler with Adaptive Time-Budget Reallocation for Time-Triggered Edge-Fog-Cloud Architectures

The lack of determinism restricts the integration of safety-critical applications into Edge–Fog–Cloud (EFC) architectures. Existing EFC schedulers are typically designed for dynamic, best-effort operation based on unmanaged resource allocation and elastic virtualization. This paradigm introduces unbounded queueing, resource contention, and timing jitter, making standard schedulers unsuitable for hard-deadline workloads. Moreover, most approaches focus on computational placement, while communication is abstracted or treated as a secondary cost term. As a result, bounded-latency routing and deterministic task execution are rarely co-optimized under a unified timing model. This paper addresses these gaps by utilizing a managed Time-Triggered Edge–Fog–Cloud (TTEFC) architecture that supports safety-critical workloads, orchestrates IEEE Time-Sensitive Networking (TSN) for local intra-domain communication, and uses IETF Deterministic Networking (DetNet) for routed inter-domain paths. On this infrastructure, a hierarchical genetic algorithm (HGA) is proposed to jointly schedule partition-to-execution-location allocation, partition execution order, inter-partition route selection, and negotiated per-partition time budgets that act as temporal boundaries for parallel partition-level optimizers. An adaptive slack reallocation operator redistributes unused temporal slack from over-satisfied partitions to budget-violating partitions, improving feasibility convergence. Experiments on synthetic DAG workloads with 100–500 tasks compare the proposed HGA against HEFT and round-robin baselines. These baselines are included as scoped external references to contextualize the end-to-end scheduling performance of the proposed method. Ablation results show that slack reallocation improves partition-budget feasibility, reaches feasible budget assignments earlier, and produces tighter budget–makespan alignment than feedback-free and static-budget variants. An automotive-characteristic DAG case study further evaluates the method on an application-oriented workload under the same timing and communication assumptions.

Omar Hekal, Josepaul Paulachan, Daniel Onwuchekwa et al. · 0 citations
Open access Aug 2026

Cloud-Network Collaborative Architecture Design and Performance Optimization Based on SRv6 and Intent-Driven Approach

Efficient cloud-network collaboration requires intelligent service orchestration, adaptive routing, and dynamic resource scheduling in distributed environments. This study proposes an intent-driven cloud-network collaborative architecture based on Segment Routing over IPv6 (SRv6). The architecture integrates intent parsing, intelligent control, and SRv6 forwarding mechanisms to achieve automated service-to-policy mapping and adaptive path orchestration. Reinforcement-learning-based routing optimization and real-time network-state feedback mechanisms are incorporated to improve resource utilization and service deployment efficiency. Experimental evaluation demonstrates significant reductions in latency and improvements in automation and resource utilization. The framework provides an effective solution for programmable networking and cloud-edge collaboration.

Kui Liu · 0 citations