Stochastic Joint Radio and Computational Resource Management for Multi-User Mobile-Edge Computing Systems (1702.00892v1)

Abstract: Mobile-edge computing (MEC) has recently emerged as a prominent technology to liberate mobile devices from computationally intensive workloads, by offloading them to the proximate MEC server. To make offloading effective, the radio and computational resources need to be dynamically managed, to cope with the time-varying computation demands and wireless fading channels. In this paper, we develop an online joint radio and computational resource management algorithm for multi-user MEC systems, with the objective as minimizing the long-term average weighted sum power consumption of the mobile devices and the MEC server, subject to a task buffer stability constraint. Specifically, at each time slot, the optimal CPU-cycle frequencies of the mobile devices are obtained in closed forms, and the optimal transmit power and bandwidth allocation for computation offloading are determined with the Gauss-Seidel method; while for the MEC server, both the optimal frequencies of the CPU cores and the optimal MEC server scheduling decision are derived in closed forms. Besides, a delay-improved mechanism is proposed to reduce the execution delay. Rigorous performance analysis is conducted for the proposed algorithm and its delay-improved version, indicating that the weighted sum power consumption and execution delay obey an $\left[O\left(1\slash V\right),O\left(V\right)\right]$ tradeoff with $V$ as a control parameter. Simulation results are provided to validate the theoretical analysis and demonstrate the impacts of various parameters.

Stochastic Joint Radio and Computational Resource Management for Multi-User MEC Systems

The paper under review presents an online algorithm for managing radio and computational resources in Multi-User Mobile-Edge Computing (MEC) systems. This work targets the optimization of MEC systems, which aim to alleviate the computational burden on mobile devices by offloading tasks to nearby MEC servers.

Problem Context and Objective

MEC is positioned as a crucial enabler for mobile devices constrained by limited computational capabilities. This necessity arises from the increasing demand imposed by computation-intensive applications like 3D modeling and gesture recognition. The primary objective of the paper is to minimize the long-term average weighted sum power consumption of the mobile devices alongside the MEC server, under the constraint of task buffer stability. The challenge lies in the dynamic management of computational resources to balance power consumption while maintaining task execution delay within acceptable bounds.

Methodology

The authors employ Lyapunov optimization to develop a low-complexity online algorithm that accomplishes joint management of radio and computational resources. Specifically, the algorithm decides on the optimal CPU-cycle frequencies for both the mobile devices and the MEC server, along with optimal transmission power and bandwidth allocation required for task offloading. The critical advantage of the proposed solution is its adaptability to real-time variations in wireless channel conditions and task arrivals, without necessitating prior knowledge of these stochastic processes.

Key Findings and Results

The algorithm guarantees an [O(1/V), O(V)] tradeoff between the average weighted sum power consumption and the execution delay. This tradeoff implies that as the control parameter $V$ increases, power consumption approaches optimal levels at the expense of higher execution delays. Conversely, smaller $V$ achieves reduced delays with increased power consumption.

Simulation results confirm the theoretical findings, revealing how parameters such as task arrival rates and MEC server computational capacity influence the power-delay tradeoff. The paper also highlights the significance of effectively balancing the compute resources between local execution and offloading tasks based on current queue lengths and channel conditions.

Implications and Future Directions

From a practical viewpoint, this research informs MEC deployment strategies by detailing how computational and radio resources should be orchestrated to balance energy efficiency and computational latency. Theoretical contributions include a deeper understanding of the implications of varying system parameters on achieving task buffer stability without compromising on power consumption.

Future research might explore distributed implementations or adaptations of this algorithm in decentralized settings where multiple MEC servers exist. Additionally, integrating mobility awareness into the resource allocation framework may yield pertinent insights, especially for scenarios involving varying user dynamics and mobility patterns.

This work underscores the potential of stochastic optimization methodologies in effectively leveraging the combined computational power of mobile devices and MEC systems, paving the way for more responsive and energy-efficient mobile application experiences.