Joint Task Offloading and Resource Allocation for Multi-Server Mobile-Edge Computing Networks (1705.00704v1)

Abstract: Mobile-Edge Computing (MEC) is an emerging paradigm that provides a capillary distribution of cloud computing capabilities to the edge of the wireless access network, enabling rich services and applications in close proximity to the end users. In this article, a MEC enabled multi-cell wireless network is considered where each Base Station (BS) is equipped with a MEC server that can assist mobile users in executing computation-intensive tasks via task offloading. The problem of Joint Task Offloading and Resource Allocation (JTORA) is studied in order to maximize the users' task offloading gains, which is measured by the reduction in task completion time and energy consumption. The considered problem is formulated as a Mixed Integer Non-linear Program (MINLP) that involves jointly optimizing the task offloading decision, uplink transmission power of mobile users, and computing resource allocation at the MEC servers. Due to the NP-hardness of this problem, solving for optimal solution is difficult and impractical for a large-scale network. To overcome this drawback, our approach is to decompose the original problem into (i) a Resource Allocation (RA) problem with fixed task offloading decision and (ii) a Task Offloading (TO) problem that optimizes the optimal-value function corresponding to the RA problem. We address the RA problem using convex and quasi-convex optimization techniques, and propose a novel heuristic algorithm to the TO problem that achieves a suboptimal solution in polynomial time. Numerical simulation results show that our algorithm performs closely to the optimal solution and that it significantly improves the users' offloading utility over traditional approaches.

Joint Task Offloading and Resource Allocation for Multi-Server Mobile-Edge Computing Networks

The paper under discussion proposes a comprehensive approach to the critical problem of joint task offloading and resource allocation in multi-cell, multi-server Mobile-Edge Computing (MEC) networks. Given the proliferation of mobile applications and the Internet of Things (IoT), MEC provides a promising paradigm by enabling cloud capabilities at the network edge, thus minimizing latency and preserving user experience continuity. The paper offers innovative solutions for optimizing user offloading gains by minimizing task completion time and energy consumption, which are central metrics in MEC environments.

Problem Formulation and Approach

The authors formulate the joint task offloading and resource allocation (JTORA) problem as a Mixed-Integer Non-Linear Program (MINLP) to maximize the system utility. Recognizing the NP-hard nature of the problem, they decompose it into a Resource Allocation (RA) problem and a Task Offloading (TO) problem. This decomposition allows the authors to address the complex joint optimization by tackling more tractable subproblems.

1. Task Offloading (TO): This problem maximizes the system utility function by determining which tasks should be offloaded and to which MEC server. It is solved by leveraging a heuristic algorithm designed to find a suboptimal solution efficiently.
2. Resource Allocation (RA): Once the decision to offload tasks is fixed, the RA problem, which includes uplink power allocation and computing resource allocation, is solved. The uplink power allocation problem is approached using quasi-convex optimization techniques, whereas the computing resource allocation problem is addressed using convex optimization techniques.

Key Contributions

• Optimization Model: The paper develops an optimization model that holistically integrates task offloading decisions with radio and computational resource allocation to maximize system utility.
• Problem Decomposition: A novel decomposition approach is proposed, breaking down the complex JTORA problem into manageable subproblems—RA and TO—each of which is further simplified into uplink power allocation and computing resource allocation.
• Heuristic Algorithm: A low-complexity heuristic algorithm is designed for the TO problem, which ensures a polynomial-time suboptimal solution. The method promises close-to-optimal performance with significantly reduced computational overhead compared to the exhaustive search method.
• Numerical Validation: Extensive simulations validate the effectiveness of the proposed solution. The heuristic algorithm not only performs closely to the optimal solution but also shows significant improvements over traditional approaches in terms of users' offloading utility.

Numerical Results

The numerical results highlight the efficacy of the proposed approach:

• Utility Performance: The heuristic algorithm consistently outperforms baseline strategies like Greedy Offloading and Joint Resource Allocation (GOJRA) and Independent Offloading and Joint Resource Allocation (IOJRA) across various metrics.
• Scalability: The solution demonstrates scalability by maintaining performance gains with increasing numbers of users and diversity in task profiles.
• Computational Efficiency: Compared to exhaustive search methods, the proposed algorithm achieves near-optimal results with substantially lower computational complexity, making it viable for real-time application in MEC systems.

Implications and Future Directions

Practically, the proposed methods can significantly enhance the efficiency of MEC systems by optimizing resource utilization and improving Quality of Experience (QoE) for users. Theoretically, the decomposition approach presents a robust framework for addressing other complex joint optimization problems in distributed computing systems.

Future research could explore several dimensions:

• Dynamic Environments: Extending the approach to dynamic environments where task profiles and network conditions vary over time.
• Multiple Antennas: Investigating the impact of deploying multiple antennas at base stations for uplink transmissions.
• Cross-Layer Optimization: Integrating the proposed methods with higher-layer protocols and load balancing mechanisms to further enhance system performance.

In summary, this paper makes a substantial contribution to the field of MEC by addressing the multifaceted problem of joint task offloading and resource allocation with innovative, efficient, and scalable solutions. The proposed methods promise to advance both the theory and practice of MEC, paving the way for more sophisticated and high-performance mobile-edge services.