Dynamic Computation Offloading for Mobile-Edge Computing with Energy Harvesting Devices (1605.05488v1)

Abstract: Mobile-edge computing (MEC) is an emerging paradigm to meet the ever-increasing computation demands from mobile applications. By offloading the computationally intensive workloads to the MEC server, the quality of computation experience, e.g., the execution latency, could be greatly improved. Nevertheless, as the on-device battery capacities are limited, computation would be interrupted when the battery energy runs out. To provide satisfactory computation performance as well as achieving green computing, it is of significant importance to seek renewable energy sources to power mobile devices via energy harvesting (EH) technologies. In this paper, we will investigate a green MEC system with EH devices and develop an effective computation offloading strategy. The execution cost, which addresses both the execution latency and task failure, is adopted as the performance metric. A low-complexity online algorithm, namely, the Lyapunov optimization-based dynamic computation offloading (LODCO) algorithm is proposed, which jointly decides the offloading decision, the CPU-cycle frequencies for mobile execution, and the transmit power for computation offloading. A unique advantage of this algorithm is that the decisions depend only on the instantaneous side information without requiring distribution information of the computation task request, the wireless channel, and EH processes. The implementation of the algorithm only requires to solve a deterministic problem in each time slot, for which the optimal solution can be obtained either in closed form or by bisection search. Moreover, the proposed algorithm is shown to be asymptotically optimal via rigorous analysis. Sample simulation results shall be presented to verify the theoretical analysis as well as validate the effectiveness of the proposed algorithm.

• The paper presents the LODCO algorithm, a low-complexity method that balances execution latency and task failure using a novel execution cost metric.
• It employs a Lyapunov optimization framework to dynamically adjust CPU frequency and transmit power based on instantaneous system states.
• Simulation results demonstrate significant reductions in execution cost and near-zero task drop ratios, confirming the algorithm’s practical benefits.

Dynamic Computation Offloading for Mobile-Edge Computing with Energy Harvesting Devices

The paper "Dynamic Computation Offloading for Mobile-Edge Computing with Energy Harvesting Devices" by Yuyi Mao, Jun Zhang, and Khaled B. Letaief, addresses the challenge of optimizing computation offloading for mobile-edge computing (MEC) systems powered by energy harvesting (EH). The primary focus is on developing an online algorithm that can dynamically balance execution latency and task failure while operating under the constraints posed by the variability of harvested energy and the computational demands of mobile applications.

Key Contributions

The paper makes several significant contributions to the field:

1. Optimization Framework: The authors introduce an execution cost metric that combines the execution delay and the penalty for task failure. This metric serves as the optimization goal for their computation offloading strategy.
2. Online Algorithm Development: The authors propose the Lyapunov optimization-based dynamic computation offloading (LODCO) algorithm. This low-complexity online algorithm determines the offloading decisions, CPU-cycle frequencies for mobile execution, and transmit power for task offloading, solely based on instantaneous side information.
3. Rigorous Theoretical Analysis: A detailed asymptotic analysis is conducted to demonstrate that the proposed algorithm is asymptotically optimal. It is shown to perform well without requiring distribution information about the computation task requests, wireless channel conditions, and EH processes.
4. Empirical Validation: The effectiveness of the LODCO algorithm is validated through simulations, which illustrate significant improvements in execution cost and task completion rates compared to several benchmark strategies.

Numerical Results

To assess the performance of the LODCO algorithm, the authors conducted extensive simulations. The results reveal that the LODCO algorithm significantly outperforms baseline strategies, particularly in scenarios with varying task arrival rates ($\rho$) and different EH rates ($P_H$). For instance, the average execution cost achieved by LODCO decreases with the EH rate, showcasing its ability to leverage renewable energy efficiently. Additionally, the LODCO algorithm maintains a near-zero task drop ratio, which is a crucial metric for user satisfaction in computation-intensive applications.

Implications and Future Directions

The implications of this research are twofold: practical and theoretical. Practically, the LODCO algorithm offers a robust solution for MEC systems with EH, ensuring high-quality user experiences by minimizing execution delays and task failures. This is particularly relevant for applications such as real-time data processing in IoT networks and mobile augmented reality services.

Theoretically, this work contributes to the understanding of dynamic resource allocation in stochastic environments. The use of Lyapunov optimization in the context of MEC and EH introduces a versatile approach that can be extended to other domains requiring online decision-making under uncertainty.

Future Developments

Future work could explore the integration of the LODCO algorithm into multi-user MEC systems, where resource sharing and user-specific constraints pose additional challenges. Additionally, extending the algorithm to account for scenarios where MEC servers have limited computational resources would be a valuable direction. Lastly, combining MEC with controllable wireless energy transfer, such as that from power beacons, could further enhance system reliability and performance by compensating for renewable energy variability.

In conclusion, the paper provides a comprehensive and effective approach to dynamic computation offloading in MEC systems with energy harvesting devices. The LODCO algorithm not only achieves asymptotically optimal performance but also is readily adaptable, offering significant practical benefits for the deployment of green and efficient MEC systems.