Dependency-aware computation-offloading and cooperative service-caching are pivotal technologies in mobile edge computing (MEC) systems. However, existing works often overlook the interdependencies among tasks/subtasks and the limited caching capacity of MEC servers, leading to suboptimal energy-time cost (ETC). In this paper, we investigate the joint optimization of computation-offloading, service-caching, and resource-allocation in an environment with multiple MEC servers, where multiple Internet of things devices (IoTDs) generate tasks composed of interdependent subtasks which are modeled as directed acyclic graphs (DAGs). Considering the constraints of limited caching capacity and computation resources at MEC servers, we formulate a mixed-integer nonlinear programming (MINLP) problem aimed at minimizing the weighted sum of task completion delay and energy consumption across all IoTDs, i.e., ETC. To solve this challenging problem, we design a scheme by integrating graph convolutional networks (GCN) with deep reinforcement learning (DRL), in which GCN extracts subtask features and dependencies among subtasks, and combines them with the features of the system environment as the state of our DRL method. We then employ an advantage actor-critic (A2C) framework with hybrid action space handling capability to perform Markov decisions and obtain optimal caching, offloading, and resource-allocation actions. Third, we conduct extensive experiments to evaluate the effectiveness of our proposed scheme. The simulations demonstrate that, compared with state-of-the-art baselines, our scheme achieves the lowest ETC in most cases, converges faster, and maintains robustness under diverse experimental conditions, including varying transmission rates, numbers of IoTDs, subtask counts, and caching capacity. Moreover, by conducting comparative analyses with relevant literature, we further confirm that our joint optimization framework has significant advantages in balancing energy efficiency and latency in the MEC environment.
- Open Access
- Article
Joint Optimization for Dependency-Aware Computation-Offloading and Cooperative Service-Caching in Edge Computing
- Yaoqian Gui,
- Fei Wang *
Author Information
Received: 23 Dec 2025 | Revised: 21 Mar 2026 | Accepted: 27 Mar 2026 | Published: 02 Jun 2026
Abstract
Keywords
mobile edge computing (MEC) | directed acyclic graph (DAG) | computation-offloading | cooperative service-caching | graph convolutional networks (GCN) | deep reinforcement learning (DRL)
References
- 1.
Mao, Y.; You, C.; Zhang, J.; et al. A Survey on Mobile Edge Computing: The Communication Perspective. IEEE Commun. Surv. Tutor. 2017, 19, 2322–2358.
- 2.
Guim, F.; Metsch, T.; Moustafa, H.; et al. Autonomous Lifecycle Management for Resource-Efficient Workload Orchestration for Green Edge Computing. IEEE Trans. Green Commun. Netw. 2022, 6, 571–582.
- 3.
Luo, R.; Jin, H.; He, Q.; et al. Cost-Effective Edge Server Network Design in Mobile Edge Computing Environment. IEEE Trans. Sustain. Comput. 2022, 7, 839–850.
- 4.
Qiu, H.; Zhu, K.; Luong, N.C.; et al. Applications of Auction and Mechanism Design in Edge Computing: A Survey. IEEE Trans. Cogn. Commun. Netw. 2022, 8, 1034–1058.
- 5.
Li, X.; Xu, Z.; Fang, F.; et al. Task Offloading for Deep Learning Empowered Automatic Speech Analysis in Mobile Edge-Cloud Computing Networks. IEEE Trans. Cloud Comput. 2023, 11, 1985–1998.
- 6.
Koganti, Y.; Sridhar, V.; Yadav, R.N.; et al. QoS-Aware Application Assignment and Resource Utilization Maximization Using AHP in Edge Computing. IEEE Internet Things J. 2025, 12, 17717–17728.
- 7.
Zhou, X.; Ge, S.; Liu, P.; et al. DAG-Based Dependent Tasks Offloading in MEC-Enabled IoT with Soft Cooperation. IEEE Trans. Mob. Comput. 2024, 23, 6908–6920.
- 8.
Deng, X.; Yang, H.; Zhang, J.; et al. Task Offloading in Internet of Vehicles: A DRL-Based Approach with Representation Learning for DAG Scheduling. IEEE Trans. Mob. Comput. 2025, 24, 5045–5060.
- 9.
Liu, H.; Huang, W.; Kim, D.I.; et al. Towards Efficient Task Offloading with Dependency Guarantees in Vehicular Edge Networks Through Distributed Deep Reinforcement Learning. IEEE Trans. Veh. Technol. 2024, 73, 13665–13681.
- 10.
Mo, C.-T.; Chen, J.-H.; Liao, W. Graph Convolutional Network Augmented Deep Reinforcement Learning for Dependent Task Offloading in Mobile Edge Computing. In Proceedings of the 2023 IEEE Wireless Communications and Networking Conference (WCNC), Glasgow, UK, 26–29 March 2023; pp. 1–6.
- 11.
Gao, M.; Shen, R.; Li, J.; et al. Computation Offloading with Instantaneous Load Billing for Mobile Edge Computing. IEEE Trans. Serv. Comput. 2022, 15, 1473–1485.
- 12.
Yan, J.; Bi, S.; Zhang, Y.J.A. Offloading and Resource Allocation with General Task Graph in Mobile Edge Computing: A Deep Reinforcement Learning Approach. IEEE Trans. Wireless Commun. 2020, 19, 5404–5419.
- 13.
Baghban, H.; Rezapour, A.; Hsu, C.-H.; et al. Edge-AI: IoT Request Service Provisioning in Federated Edge Computing Using Actor-Critic Reinforcement Learning. IEEE Trans. Eng. Manag. 2024, 71, 12519–12528.
- 14.
Goudarzi, S.; Soleymani, S.A.; Anisi, M.H.; et al. Optimizing UAV-Assisted Vehicular Edge Computing with Age of Information: An SAC-Based Solution. IEEE Internet Things J. 2025, 12, 4555–4569.
- 15.
Zhou, J.; Zhao, Y.; Zhao, L.; et al. Adaptive Task Offloading with Spatiotemporal Load Awareness in Satellite Edge Computing. IEEE Trans. Netw. Serv. Manag. 2024, 11, 5311–5322.
- 16.
Zhang, J.; Zhang, G.; Bao, X.; et al. Dependent Application Offloading in Edge Computing. IEEE Trans. Cloud Comput. 2023, 11, 3439–3451.
- 17.
Liu, J.; Ren, J.; Zhang, Y.; et al. Efficient Dependent Task Offloading for Multiple Applications in MEC-Cloud System. IEEE Trans. Mob. Comput. 2023, 22, 2147–2162.
- 18.
Chen, J.; Yang, Y.; Wang, C.; et al. Multitask Offloading Strategy Optimization Based on Directed Acyclic Graphs for Edge Computing. IEEE Internet Things J. 2022, 12, 9367–9378.
- 19.
Liao, Z.; Cheng, S.; Zhang, J.; et al. GpDB: A Graph-partition Based Storage Strategy for DAG-Blockchain in Edge-cloud IIoT. IEEE Trans. Ind. Inf. 2025, 21, 5790–5798.
- 20.
Wang, Z.; Goudarzi, M.; Buyya, R. TF-DDRL: A Transformer-Enhanced Distributed DRL Technique for Scheduling IoT Applications in Edge and Cloud Computing Environments. IEEE Trans. Serv. Comput. 2025, 18, 1039–1053.
- 21.
Goudarzi, M.; Palaniswami, M.; Buyya, R. A Distributed Deep Reinforcement Learning Technique for Application Placement in Edge and Fog Computing Environments. IEEE Trans. Mob. Comput. 2023, 22, 2491–2505.
- 22.
Goudarzi, M.; Rodriguez, M.A.; Sarvi, M.; et al. µ-DDRL: A QoS-Aware Distributed Deep Reinforcement Learning Technique for Service Offloading in Fog Computing Environments. IEEE Trans. Serv. Comput. 2024, 17, 47–59.
- 23.
Chen, Z.; Wang, F.; Zhang, X. Joint Optimization for Cooperative Service-Caching, Computation-Offloading, and Resource-Allocations over EH/MEC-based Ultra-Dense Mobile Networks. In Proceedings of the 2023 IEEE International Conference on Communications (ICC), Rome, Italy, 28 May–01 June 2023; pp. 716–722.
- 24.
Yan, H.; Xu, X.; Bilal, M.; et al. Customer Centric Service Caching for Intelligent Cyber-Physical Transportation Systems with Cloud-Edge Computing Leveraging Digital Twins. IEEE Trans. Consum. Electron. 2024, 70, 1787–1797.
- 25.
Chen, Z.; Liang, J.; Yu, Z.; et al. Resilient Collaborative Caching for Multi-Edge Systems with Robust Federated Deep Learning. IEEE Trans. Netw. 2025, 33, 654–669.
- 26.
Zeng, J.; Zhou, X.; Li, K. Resource-Efficient Joint Service Caching and Workload Scheduling in Ultra-Dense MEC Networks: An Online Approach. IEEE Trans. Netw. Serv. Manag. 2025, 22, 1788–1800.
- 27.
Zhou, H.; Zhang, Z.; Li, D.; et al. Joint Optimization of Computing Offloading and Service Caching in Edge Computing-Based Smart Grid. IEEE Trans. Cloud Comput. 2023, 11, 1122–1132.
- 28.
Liu, L.; Chen, Z. Joint Optimization of Multiuser Computation Offloading and Wireless-Caching Resource Allocation with Linearly Related Requests in Vehicular Edge Computing System. IEEE Internet Things J. 2024, 11, 1534–1547.
- 29.
Li, Y.; Zhu, X.; Li, N.; et al. Collaborative Content Caching and Task Offloading in Multi-Access Edge Computing. IEEE Trans. Veh. Technol. 2023, 72, 5367–5372.
- 30.
Zhao, Y.; Liu, C.; Hu, X.; et al. Quek. Joint Content Caching, Service Placement, and Task Offloading in UAV-Enabled Mobile Edge Computing Networks. IEEE J. Sel. Areas Commun. 2025, 43, 51–63.
- 31.
Wei, X.; Cai, L.; Wei, N.; et al. Joint UAV Trajectory Planning, DAG Task Scheduling, and Service Function Deployment Based on DRL in UAV-Empowered Edge Computing. IEEE Internet Things J. 2023, 10, 12826–12838.
- 32.
Zhao, L.; Zhao, Z.; Hawbani, A.; et al. Dynamic Caching Dependency-Aware Task Offloading in Mobile Edge Computing. IEEE Trans. Comput. 2025, 74, 1510–1523.
- 33.
Sun, Y.; Zhang, X. A2C Learning for Tasks Segmentation with Cooperative Computing in Edge Computing Networks. In Proceedings of the 2022 IEEE Global Communications Conference (GLOBECOM), Rio de Janeiro, Brazil, 4–8 December 2022; pp. 2236–2241.
- 34.
Lu, J.; Yang, J.; Li, S.; et al. A2C-DRL: Dynamic Scheduling for Stochastic Edge-Cloud Environments Using A2C and Deep Reinforcement Learning. IEEE Internet Things J. 2024, 11, 16915–16927.
- 35.
Wang, J.; Wang, F.; Zhang, X.; et al. Joint Optimizations for Double-IRS’ Cooperative Positioning and Beamforming Over Massive-MIMOAP Based 6G Secure Mobile Wireless Networks. In Proceedings of the 2024 IEEE Global Communications Conference (GLOBECOM), Cape Town, South Africa, 8–12 December 2024; pp. 1–6.
- 36.
Tong, R.; Guo, S.; Qiu, X.; et al. GCN and DRL Based on Dependent Task Offloading Mechanism in Edge Computing. Digit. Commun. Netw. 2025, 12, 397–404.
- 37.
Wang, Y.; Chen, J.; Wu, Z.; et al. Efficient Task Migration and Resource Allocation in Cloud–Edge Collaboration: A DRL Approach with Learnable Masking. Alex. Eng. J. 2025, 111, 107–122.
- 38.
Li, H.; Li, P.; Assis, K.R.D.; et al. NetMind+: Adaptive Baseband Function Placement with GCN Encoding and Incremental Maze-Solving DRL for Dynamic and Heterogeneous RANs. IEEE Trans. Netw. Serv. Manag. 2025, 22, 3419–3432.
This work is licensed under a Creative Commons Attribution 4.0 International License.
Suite 4002 Level 4, 447 Collins Street, Melbourne, Victoria 3000, Australia
General Inquiries: info@sciltp.com