Mobile Edge Computing Server Research Articles

Enhancing M-Government service adoption in Saudi Arabia: the role of mobile edge computing in moderating technology acceptance

Purpose Mobile edge computing (MEC) services have long been used by private enterprises in Saudi Arabia with considerable success; however, there has been a stark lack of insight into how these services can be used to improve mobile government (M-Government) services for KSA citizens. This study aims to bridge this gap by integrating MEC with an enhanced version of the technology acceptance model (TAM) and examining its effects on user behavior and acceptance. Design/methodology/approach A closed-ended survey was administered to 1,500 people, and the responses were analyzed using sophisticated advanced statistical techniques to test an expanded TAM, using a quantitative method that uses structural equation modeling to validate the proposed model and hypotheses. Findings This study reveals that MEC significantly influences users’ intentions about using M-Government services and their tolerance for new technology adoption. Specifically, service cost and social influence are positively linked with end users’ intention to adopt M-Government services. Originality/value The novelty and contribution of this paper to existing literature are in highlighting the pivotal role of MEC in transforming public sector service delivery through technology. This study not only supports the adoption of M-Government services to enhance social welfare but also demonstrates and concludes some practical and theoretical ramifications of MEC service adoption.

Joint Computation Offloading and Trajectory Optimization for Edge Computing UAV: A KNN-DDPG Algorithm

Unmanned aerial vehicles (UAVs) are widely used to improve the coverage and communication quality of wireless networks and assist mobile edge computing (MEC) due to their flexible deployments. However, the UAV-assisted MEC systems also face challenges in terms of computation offloading and trajectory planning in the dynamic environment. This paper employs deep reinforcement learning to jointly optimize the computation offloading and trajectory planning for UAV-assisted MEC system. Specifically, this paper investigates a general scenario where multiple pieces of user equipment (UE) offload tasks to a UAV equipped with a MEC server to collaborate on a complex job. By fully considering UAV and UE movement, computation offloading ratio, and blocked relations, a joint computation offloading and trajectory optimization problem is formulated to minimize the maximum computational delay. Due to the non-convex nature of the problem, it is converted into a Markov decision process, and solved by the deep deterministic policy gradient (DDPG) algorithm. To enhance the exploration capability and stability of DDPG, the K-nearest neighbor (KNN) algorithm is employed, namely KNN-DDPG. Moreover, the prioritized experience replay algorithm, where the constant learning rate is replaced by the decaying learning rate, is utilized to enhance the converge. To validate the effectiveness and superiority of the proposed algorithm, KNN-DDPG is compared with the benchmark DDPG algorithm. Simulation results demonstrate that KNN-DDPG can converge and achieve 3.23% delay reduction compared to DDPG.

Decision Transformer-Based Efficient Data Offloading in LEO-IoT.

Recently, the Internet of Things (IoT) has witnessed rapid development. However, the scarcity of computing resources on the ground has constrained the application scenarios of IoT. Low Earth Orbit (LEO) satellites have drawn people's attention due to their broader coverage and shorter transmission delay. They are capable of offloading more IoT computing tasks to mobile edge computing (MEC) servers with lower latency in order to address the issue of scarce computing resources on the ground. Nevertheless, it is highly challenging to share bandwidth and power resources among multiple IoT devices and LEO satellites. In this paper, we explore the efficient data offloading mechanism in the LEO satellite-based IoT (LEO-IoT), where LEO satellites forward data from the terrestrial to the MEC servers. Specifically, by optimally selecting the forwarding LEO satellite for each IoT task and allocating communication resources, we aim to minimize the data offloading latency and energy consumption. Particularly, we employ the state-of-the-art Decision Transformer (DT) to solve this optimization problem. We initially obtain a pre-trained DT through training on a specific task. Subsequently, the pre-trained DT is fine-tuned by acquiring a small quantity of data under the new task, enabling it to converge rapidly, with less training time and superior performance. Numerical simulation results demonstrate that in contrast to the classical reinforcement learning approach (Proximal Policy Optimization), the convergence speed of DT can be increased by up to three times, and the performance can be improved by up to 30%.

Integrated trajectory optimization for UAV-enabled wireless powered MEC system with joint energy consumption and AoI minimization

This paper studies an unmanned aerial vehicle (UAV)-enabled wireless powered mobile edge computing (MEC) system, where a UAV, equipped with RF Chains and MEC servers, can sustainably provide wireless energy for charging Internet of Things (IoT) devices and executing computing tasks from these devices while hovering at designated hover points. Our goal is to minimize the weighted sum of energy consumption and Age of information (AoI) in this system, which depended on the UAV’s hovering time at designated points and its flying time. To achieve this, we jointly optimize the deployment of hover points and the visiting order of these points by the UAV. It is NP hard and mixed-integer non-convex which is difficult to solve by traditional methods. To tackle this problem, we present a trajectory optimization algorithm for joint energy consumption and AoI (TOJEA), which consists of two phases. In the first phase, an Equilibrium Optimizer (EO) algorithm with a variable individual size via its coding and updating strategies, in which each particle (individual) with its concentration (position) represents a target solution i.e. the whole deployment of hover points, is proposed to optimize the number and locations of hover points. Based on the deployment of hover points, a low-complexity greedy algorithm is adopted in the second stage to generate the optimal visiting order for the UAV. Experimental results demonstrate that TOJEA outperforms other algorithms on ten instances with up to 400 IoT devices.

DDQN-based online computation offloading and application caching for dynamic edge computing service management

Multi-access Edge Computing (MEC) reduces task service latency and energy consumption by offloading computing tasks to MEC servers. However, constrained by the limited bandwidth and computing resources, MEC servers often cannot parallelly process all computing tasks. Simultaneously, the high dynamism of service popularity necessitates MEC servers to dynamically update cached applications, under ensuring compliance with storage resource constraints and the system cache updating cost budget for service providers. In response to the above two issues, this paper firstly formulates computation offloading and application caching as a dual-timescale decision optimization problem, aiming to minimize the average service latency for users by obtaining optimal offloading decision, cache decision, transmission bandwidth, and computing resource. Then, we propose a Deep Reinforcement Learning (DRL)-based two-stage online computation offloading and application caching (DTSO2C) algorithm, effectively stabilizing application cache update costs and enhancing Quality of Service (QoS) for users. Furthermore, we utilize convex optimization algorithms to derive the optimal communication bandwidth and computing resource allocation strategy, further reducing the average service latency for users. Simulation results demonstrate that the DTSO2C algorithm outperforms the compared algorithms, achieving an average reduction in service latency of 66.2%, with an average cache update cost of only 0.15 USD per time frame.

Load-balanced offloading of multiple task types for mobile edge computing in IoT

The continuous development of mobile networks poses new challenges for end devices with limited computing power. Mobile or multi-access edge computing (MEC) has been proposed for providing the computing resources close to the end devices that need them. However, in real network environments, MEC servers also have limited computing resources that need to be shared among many devices and efficient resource allocation is critical to ensure that the limited resources are optimally used. In view of this, we propose the Balanced Offload for Multi-type Tasks (BOMT) algorithm. The tasks to be offloaded are first prioritized according to their type, size and maximum tolerable delay; then different offloading algorithms are executed for different priority tasks according to the level of the priority and the current load on the MEC server. Following which, the optimal offloading policy is determined iteratively. Simulation results show that BOMT can effectively reduce system delay, increase user coverage and offload task completion rates.

Generative AI-Enabled Energy-Efficient Mobile Augmented Reality in Multi-Access Edge Computing

This paper proposes a novel offloading and super-resolution (SR) control scheme for energy-efficient mobile augmented reality (MAR) in multi-access edge computing (MEC) using SR as a promising generative artificial intelligence (GAI) technology. Specifically, SR can enhance low-resolution images into high-resolution versions using GAI technologies. This capability is particularly advantageous in MAR by lowering the bitrate required for network transmission. However, this SR process requires considerable computational resources and can introduce latency, potentially overloading the MEC server if there are numerous offload requests for MAR services. In this context, we conduct an empirical study to verify that the computational latency of SR increases with the upscaling level. Therefore, we demonstrate a trade-off between computational latency and improved service satisfaction when upscaling images for object detection, as it enhances the detection accuracy. From this perspective, determining whether to apply SR for MAR, while jointly controlling offloading decisions, is challenging. Consequently, to design energy-efficient MAR, we rigorously formulate analytical models for the energy consumption of a MAR device, the overall latency and the MAR satisfaction of service quality from the enforcement of the service accuracy, taking into account the SR process at the MEC server. Finally, we develop a theoretical framework that optimizes the computation offloading and SR control problem for MAR clients by jointly optimizing the offloading and SR decisions, considering their trade-off in MAR with MEC. Finally, the performance evaluation indicates that our proposed framework effectively supports MAR services by efficiently managing offloading and SR decisions, balancing trade-offs between energy consumption, latency, and service satisfaction compared to benchmarks.

Optimizing Energy and Delay in Task Offloading for Connected Vehicles with Proximal Policy Optimization

Electric-powered intelligent connected vehicles are becoming the pivotal point of the automotive industry. As vehicles integrate more applications, the computation tasks they generate increase substantially. Cloud servers cannot handle these tasks promptly, and current electric vehicles (EVs) have limited energy and computing resources. Multi-Access edge computing (MEC) performs various activities in proximity to the vehicles, resulting in decreased latency and the preservation of EV battery power. However, MEC servers have finite processing resources and may be unable to satisfy the required latency restrictions. We propose a task offloading scheme to optimize the allocation of computational resources from roadside servers across several EVs. We develop a mathematical model to optimize both computation latency and EV energy, represented as a Markov decision process (MDP). To address this, we employ the deep reinforcement learning-proximal policy optimization (DRL-PPO) algorithm. The implementation of our mathematical model, which is based on an MDP, together with the use of the DRL-PPO algorithm, showcases notable decreases in both energy consumption and latency when compared to alternative benchmark deep reinforcement learning (DRL) approaches.

E-CARGO-based dynamic weight offload strategy with resource contention mitigation for edge networks

With the widespread use of Mobile Edge Computing (MEC) in smart manufacturing systems in Industrial Internet of Things (IIoT) and 5G networks, determining how to efficiently offload computing tasks has become a hot research area. The Role-Based Collaboration (RBC) Environments-Classes, Agents, Roles, Groups, and Objects (E-CARGO) model is introduced to comprehensively manage MEC servers and user computation tasks in edge network environments, thereby improving the effectiveness and performance of task offloading in smart manufacturing systems. To begin with, latency and energy consumption are important indicators for evaluating the offloading effect. A pre-allocation algorithm based on user latency tolerance is proposed to dynamically adjust the latency-energy consumption weighting factor to optimize system resource allocation for real-time adjustment of offloading decisions. Second, the Group Role Assignment of Agent Role Conflicts (GRACAR) model based on E-CARGO is extended, along with a dynamic weighting of the GRACAR (GRACAR-DW) model and formal modeling. By introducing resource contention constraints, the resource contention caused by excessive task data offloading to the same MEC server is proactively mitigated. Finally, a Gurobi solution based on Mixed-Integer Linear Programming (MILP) is developed to help validate and synthesize the proposed model. Simulation results show that the strategy considerably enhances the MEC system's overall performance in terms of latency and energy consumption while also providing new ideas and technological support for offloading decisions in edge networks.

Optimal resource management for multi-access edge computing without using cross-layer communication

We consider a Multi-access Edge Computing (MEC) system with a set of users, a base station (BS) with an attached MEC server, and a cloud server. The users can serve the service requests locally or can offload them to the BS which in turn can serve a subset of the offloaded requests at the MEC and can forward the requests to the cloud server. The user devices and the MEC server can be dynamically configured to serve different classes of services. The service requests offloaded to the BS incur offloading costs and those forwarded to the cloud incur additional costs; the costs could represent service charges or delays. Aggregate cost minimization subject to stability warrants optimal service scheduling and offloading at the users and the MEC server, at their application layers, and optimal uplink packet scheduling at the users’ MAC layers. Classical back-pressure (BP) based solutions entail cross-layer message exchange, and hence are not viable. We propose virtual queue-based drift-plus-penalty algorithms that are throughput optimal, and achieve the optimal delay arbitrarily closely but do not require cross-layer communication. We first consider an MEC system without local computation, and subsequently, extend our framework to incorporate local computation also. We demonstrate that the proposed algorithms offer almost the same performance as BP based algorithms. These algorithms contain tuneable parameters that offer a trade off between queue lengths at the users and the BS and the offloading costs.

Truthful mechanism for joint resource allocation and task offloading in mobile edge computing

In the context of mobile edge computing (MEC), the delay-sensitive tasks can achieve real-time data processing and analysis by offloading to the MEC servers. The objective is maximizing social welfare in an auction-based model. However, the distances between mobile devices and access points lead to differences in energy consumption. Unfortunately, existing works have not considered both maximizing social welfare and minimizing energy consumption. Motivated by this, we address the problem of joint resource allocation and task offloading in MEC, with heterogeneous MEC servers providing multiple types of resources for mobile devices (MDs) to perform tasks remotely. We split the problem into two sub-problems: winner determination and offloading decision. The first sub-problem determines winners granted the ability to offload tasks to maximize social welfare. The second sub-problem determines how to offload tasks among the MEC servers to minimize energy consumption. In the winner determination problem, we propose a truthful algorithm that drives the system into equilibrium. We then show the approximate ratios for single and multiple MEC servers. In the offloading decision problem, we propose an approximation algorithm. We then show it is a polynomial-time approximation scheme for a single MEC server. Experiment results show that our proposed mechanism finds high-quality solutions in changing mobile environments.

Joint optimization for service-caching, computation-offloading, and UAVs flight trajectories over rechargeable UAV-aided MEC using hierarchical multi-agent deep reinforcement learning

Due to the high mobility, high chance of line-of-sight (LoS) transmission, and flexible deployment, unmanned aerial vehicles (UAVs) have been used as mobile edge computing (MEC) servers to provide ubiquitous computation services to mobile users (MUs). However, the limited energy storage, caching capacity, and computation resources of UAVs bring new challenges for UAV-aided MEC, e.g., how to recharge UAVs and how to jointly optimize service-caching, computation-offloading, and UAVs flight trajectories. To overcome the above-mentioned difficulties, in this paper we study the joint optimization for service-caching, computation-offloading, and UAVs flight trajectories for UAV-aided MEC, where multiple rechargeable UAVs cooperatively provide MEC services to a number of MUs. First, we formulate an energy minimization problem to minimize all MUs' energy consumptions by taking into account the mobility of MUs and the energy replenishment of UAVs. Then, using the hierarchical multi-agent deep reinforcement learning (HMDRL), we develop a two-timescale based joint service-caching, computation-offloading, and UAVs flight trajectories scheme, called HMDRL-Based SCOFT. Using HMDRL-Based SCOFT, we derive UAVs' service-caching policies in each time frame, and then derive UAVs flight trajectories and MUs' computation-offloading in each time slot. Finally, we validate and evaluate the performances of our proposed HMDRL-Based SCOFT scheme through extensive simulations, which show that our developed scheme outperforms the other baseline schemes to converge faster and greatly reduce MUs' energy consumptions.

A truthful randomized mechanism for task allocation with multi-attributes in mobile edge computing

Mobile Edge Computing (MEC) aims at decreasing the response time and energy consumption of running mobile applications by offloading the tasks of mobile devices (MDs) to the MEC servers located at the edge of the network. The demands are multi-attribute, where the distances between MDs and access points lead to differences in required resources and transmission energy consumption. Unfortunately, the existing works have not considered both task allocation and energy consumption problems. Motivated by this, this paper considers the problem of task allocation with multi-attributes, where the problem consists of the winner determination and offloading decision problems. First, the problem is formulated as the auction-based model to provide flexible service. Then, a randomized mechanism is designed and is truthful in expectation. This drives the system into an equilibrium where no MD has incentives to increase the utility by declaring an untrue value. In addition, an approximation algorithm is proposed to minimize remote energy consumption and is a polynomial-time approximation scheme. Therefore, it achieves a tradeoff between optimality loss and time complexity. Simulation results reveal that the proposed mechanism gets the near-optimal allocation. Furthermore, compared with the baseline methods, the proposed mechanism can effectively increase social welfare and bring higher revenue to edge server providers.

Energy minimization for IRS-and-UAV-assisted mobile edge computing

Intelligent reconfigurable surface (IRS) is an emerging technology for the enhancement of spectrum and energy efficiency. We propose a novel IRS-and-unmanned aerial vehicle (UAV)-Assisted mobile edge computing (MEC) framework, where a MEC server installing on an UAV to facilitate task calculations by mobile users (MUs), and an IRS modulates channels between MUs and the UAV. Non-orthogonal multiple access (NOMA) is employed for further improving system-wide spectral efficiency. There are needs for joint optimization of multiple parameters, e.g., the task partition parameters and the transmit power of all MUs, the reflection coefficient matrix of the IRS and the movement trajectory of the UAV, and such needs raises the challenge of minimizing the long-term total energy consumption of all MUs while satisfying required transmission rate and task completion delay. We divide optimization tasks into two sub-problems and propose specific solutions respectively, i.e., relevant decisions about the UAV and MUs to be solved by deep reinforcement learning (DRL); and the reflection coefficient matrix of the IRS to be solved by block coordinate descent (BCD). A series of experiments have verified the effectiveness of the proposed communication techniques and optimization algorithms. Simulation results demonstrate that (1) NOMA-IRS technique achieves better energy efficacy compared to the cases where random IRS or no IRS is deployed and the conventional orthogonal multiple access (OMA) technique with IRS. (2) our proposed deep deterministic policy gradient (DDPG)-BCD algorithm outperforms other four benchmark algorithms in solving the complex and dynamic optimization problem.

TRUST-ME: Trust-Based Resource Allocation and Server Selection in Multi-Access Edge Computing

Multi-access edge computing (MEC) has attracted the interest of the research and industrial community to support Internet of things (IoT) applications by enabling efficient data processing and minimizing latency. This paper presents significant contributions toward optimizing the resource allocation and enhancing the decision-making process in edge computing environments. Specifically, the TRUST-ME model is introduced, which consists of multiple edge servers and IoT devices, i.e., users, with varied computing tasks offloaded to the MEC servers. A utility function was designed to quantify the benefits in terms of latency and cost for the IoT device while utilizing the MEC servers’ computing capacities. The core innovation of our work is a novel trust model that was designed to evaluate the IoT devices’ confidence in MEC servers. This model integrates both direct and indirect trust and reflects the trustworthiness of the servers based on the direct interactions and social feedback from other devices using the same servers. This dual trust approach helps with accurately gauging the reliability of MEC services and ensuring more informed decision making. A reinforcement learning framework based on the optimistic Q-learning with an upper confidence bounds action selection algorithm enables the IoT devices to autonomously select a MEC server to process their computing tasks. Also, a multilateral bargaining model is proposed for fair resource allocation of the MEC servers’ computing resources to the users while accounting for their computing demands. Numerical simulations demonstrated the operational effectiveness, convergence, and scalability of the TRUST-ME model, which was validated through real-world scenarios and comprehensive comparative evaluations against existing approaches.

Optimization for total energy consumption of drone inspection based on distance-constrained capacitated vehicle routing problem: A study in wind farm

Facing large-scale, intensive offshore wind farms in wide and deep sea areas, total inspection energy consumption (TIEC) optimization is a key issue to be addressed in unmanned aerial vehicle (UAV) inspection. This paper focuses on the TIEC optimization of UAV automatic inspection of offshore wind farms based on the assistance of mobile edge computing (MEC). To tackle this issue, a system model using a UAV to inspect offshore wind farms automatically and an energy consumption model of UAV assisted by MEC servers and near-earth orbiting satellites for inspecting wind turbines are proposed. Based on this, a combinatorial optimization problem based on the distance-constrained capacitated vehicle routing problem (DCVRP) for minimizing the TIEC of UAV inspection was constructed. To solve the optimization problem, an improved hybrid heuristic algorithm is proposed based on the K-means clustering, smallest enclosing circle (SEC), and Lin-Kernighan-Helsgaun (LKH-3) algorithms. The algorithm solves the optimum location and number of automated UAV airports and minimizes the TIEC for UAV inspection. The simulation experimental results show that in an offshore wind farm with 86 wind turbines, compared with other existing schemes, this scheme requires only four automated UAV airfields and can save at most 51.8% of the TIEC.

Task Offloading and Execution in Edge-Cloud Collaborative Network Using Genetic Algorithm

Mobile Edge Computing (MEC) system is now one of the most emerging sectors in wireless communication. It provides computation capability near the edge users and reduces the dependency on the Internet Cloud (IC). The edge users offload the task to the MEC server due to the lack of computing resources for executing resource-hungry applications. MEC servers can serve faster as they are quite close to the edge users but have limited computation resources compared to IC. On the other hand, IC has abundant computation resources but offloading to IC will add extra latency to complete tasks execution. In this scenario, an MEC has to decide intelligently which tasks should be executed by itself and which should be sent to IC. In this paper, we addressed this issue of task assignment and execution either on MEC or on IC and formulated an optimization problem to reduce the task processing latency. The problem is Integer Linear and NP-Complete, thus having higher time complexity. In this regard, we proposed a Genetic algorithm-based Meta-Heuristic method to solve the task completion problem considering the delay constraint of the user. Simulation results of the proposed method indicated improvement in terms of successful task execution, task turnaround time, and better Quality of experience (QoE) compared to the state-of-the-art works. GUB JOURNAL OF SCIENCE AND ENGINEERING, Vol 9(1), 2022 P 42-51

Enhanced Wireless Communication Optimization with Neural Networks, Proximal Policy Optimization and Edge Computing for Latency and Energy Efficiency

This research proposes a novel approach for efficient resource allocation in wireless communication systems. It combines dynamic neural networks, Proximal Policy Optimization (PPO), and Edge Computing Orchestrator (ECO) for latency-aware and energy-efficient resource allocation. The proposed system integrates multiple components, including a dynamic neural network, PPO, ECO, and a Mobile Edge Computing (MEC) server. The experimental methodology involves utilizing the NS-3 simulation platform to assess latency and energy efficiency in resource allocation within a wireless communication network, incorporating an ECO, MEC server, and dynamic task scheduling algorithms. It demonstrates a holistic and adaptable approach to resource allocation in dynamic environments, showcasing a notable reduction in latency for devices and tasks. Latency values range from 5 to 20 milliseconds, with corresponding resource utilization percentages varying between 80% and 95%. Additionally, energy-efficient resource allocation demonstrates a commendable reduction in energy consumption, with measured values ranging from 10 to 30 watts, coupled with efficient resource usage percentages ranging from 70% to 85%. These outcomes validate the efficacy of achieving both latency-aware and energy-efficient resource allocation for enhanced wireless communication systems. The proposed system has broad applications in healthcare, smart cities, IoT, real-time analytics, autonomous vehicles, and augmented reality, offering a valuable solution to optimize energy consumption, reduce latency, and enhance system efficiency in these industries.

Dependent Task Offloading and Resource Allocation via Deep Reinforcement Learning for Extended Reality in Mobile Edge Networks

Extended reality (XR) is an immersive technology widely applied in various fields. Due to the real-time interaction required between users and virtual environments, XR applications are highly sensitive to latency. Furthermore, handling computationally intensive tasks on wireless XR devices leads to energy consumption, which is a critical performance constraint for XR applications. It has been noted that the XR task can be decoupled to several subtasks with mixed serial–parallel relationships. Furthermore, the evaluation of XR application performance involves both subjective assessments from users and objective evaluations, such as of energy consumption. Therefore, in edge computing environments, ways to integrate task offloading for XR subtasks to meet users’ demands for XR applications is a complex and challenging issue. To address this issue, this paper constructs a wireless XR system based on mobile edge computing (MEC) and conducts research on the joint optimization of multi-user communication channel access and task offloading. Specifically, we consider the migration of partitioned XR tasks to MEC servers and formulate a joint optimization problem for communication channel access and task offloading. The objective is to maximize the ratio of quality of experience (QoE) to energy consumption while meeting the user QoE requirements. Subsequently, we introduce a deep reinforcement learning-based algorithm to address this optimization problem. The simulation results demonstrate the effectiveness of this algorithm in meeting user QoE demands and improving energy conversion efficiency, regardless of the XR task partitioning strategies employed.

DNN acceleration in vehicle edge computing with mobility-awareness: A synergistic vehicle–edge and edge–edge framework

In recent years, vehicular networks have seen a proliferation of applications and services such as image tagging, lane detection, and speech recognition. Many of these applications rely on Deep Neural Networks (DNNs) and demand low-latency computation. To meet these requirements, Vehicular Edge Computing (VEC) has been introduced to augment the abundant computation capacity of vehicular networks to complement limited computation resources on vehicles. Nevertheless, offloading DNN tasks to MEC (Multi-access Edge Computing) servers effectively and efficiently remains a challenging topic due to the dynamic nature of vehicular mobility and varying loads on the servers. In this paper, we propose a novel and efficient distributed DNN Partitioning And Offloading (DPAO), leveraging the mobility of vehicles and the synergy between vehicle–edge and edge–edge computing. We exploit the variations in both computation time and output data size across different layers of DNN to make optimized decisions for accelerating DNN computations while reducing the transmission time of intermediate data. In the meantime, we dynamically partition and offload tasks between MEC servers based on their load differences. We have conducted extensive simulations and testbed experiments to demonstrate the effectiveness of DPAO. The evaluation results show that, compared to offloaded all tasks to MEC server, DPAO reduces the latency of DNN tasks by 2.4x. DPAO with queue reservation can further reduce the task average completion time by 10%.