Offloading Decision Algorithm Research Articles

Knowledge-Driven Service Offloading Decision for Vehicular Edge Computing: A Deep Reinforcement Learning Approach

The smart vehicles construct Internet of Vehicle (IoV), which can execute various intelligent services. Although the computation capability of a vehicle is limited, multi-type of edge computing nodes provide heterogeneous resources for intelligent vehicular services. When offloading the complex service to the vehicular edge computing node, the decision for its destination should be considered according to numerous factors. This paper mostly formulate the offloading decision as a resource scheduling problem with single or multiple objective function and constraints, where some customized heuristics algorithms are explored. However, offloading multiple data dependence tasks in a complex service is a difficult decision, as an optimal solution must understand the resource requirement, the access network, the user mobility, and importantly the data dependence. Inspired by recent advances in machine learning, we propose a knowledge driven (KD) service offloading decision framework for IoV, which provides the optimal policy directly from the environment. We formulate the offloading decision for the multiple tasks as a long-term planning problem, and explore the recent deep reinforcement learning to obtain the optimal solution. It can scruple the future data dependence of the following tasks when making decision for a current task from the learned offloading knowledge. Moreover, the framework supports the pre-training at the powerful edge computing node and continually online learning when the vehicular service is executed, so that it can adapt the environment changes and can learn policy that are sensible in foresight. The simulation results show that KD service offloading decision converges quickly, adapts to different conditions, and outperforms a greedy offloading decision algorithm.

Multi-User and Multi-Task Offloading Decision Algorithms Based on Imbalanced Edge Cloud

The mobile edge computing (MEC) technology can provide mobile users (MU) with high reliability and low time-delay computing and communication services. The imbalanced edge cloud deployment can better adapt to the non-uniform spatial-time distribution of tasks and reduce the deployment cost of edge cloud servers. For multi-user and multi-task offloading decision based on the imbalanced edge cloud, a new offloading cost criteria, based on the tradeoff among time delay-energy consumption-cost, is designed to quantify the user experience of task offloading and to be the optimization target of offloading decision. Both the optimization problems of minimizing the sum offloading costs for all MUs (efficiency-based) and minimizing the maximal offloading cost per MU (fairness-based) are discussed. Efficiency-based offloading decision algorithm [centralized greedy algorithm (CGA) and modified greedy algorithm (MGA)] and fairness-based offloading decision algorithm [fairness-based greedy algorithm (FGA)] are proposed, respectively, and the performance bounds of the algorithm are analyzed. The simulation results show that the offloading cost of the MGA is lower than the CGA, the efficiency of resource utilization of the CGA is higher than that of the FGA, and the fairness of the FGA is stronger than that of the CGA.

Energy-Aware Computation Offloading of IoT Sensors in Cloudlet-Based Mobile Edge Computing.

Mobile edge computing is proposed as a promising computing paradigm to relieve the excessive burden of data centers and mobile networks, which is induced by the rapid growth of Internet of Things (IoT). This work introduces the cloud-assisted multi-cloudlet framework to provision scalable services in cloudlet-based mobile edge computing. Due to the constrained computation resources of cloudlets and limited communication resources of wireless access points (APs), IoT sensors with identical computation offloading decisions interact with each other. To optimize the processing delay and energy consumption of computation tasks, theoretic analysis of the computation offloading decision problem of IoT sensors is presented in this paper. In more detail, the computation offloading decision problem of IoT sensors is formulated as a computation offloading game and the condition of Nash equilibrium is derived by introducing the tool of a potential game. By exploiting the finite improvement property of the game, the Computation Offloading Decision (COD) algorithm is designed to provide decentralized computation offloading strategies for IoT sensors. Simulation results demonstrate that the COD algorithm can significantly reduce the system cost compared with the random-selection algorithm and the cloud-first algorithm. Furthermore, the COD algorithm can scale well with increasing IoT sensors.

Signal to Noise Ratio Based Wi-Fi Offloading Decision Algorithm in Vehicular Networks

Nowadays, the rapid increase in mobile data demand is faced by the existing mobile networks. It becomes clear that the existing cellular infrastructure, along with its next generation upgrades is unable to meet the requirements of the users to provide them a required level of throughput for their services. This paper provides an overview of the different standards and requirements in Wi-Fi offloading system. In Het-Nets, multi Radio Access Technology may be operating jointly with the cellular network. This scheme is aimed at evaluating the performance of incorporating Wi-Fi Access Points (APs) along with the cellular network. The decision is based on the selection of best Wi-Fi AP for offloading the data, for which the Signal to Noise Ratio value at the user location is more than the threshold value in vehicular environment. At the end, the numerical results evaluate the performance of presented scheme which justify its application in vehicular networks.

Spatial and Temporal Computation Offloading Decision Algorithm in Edge Cloud-Enabled Heterogeneous Networks

A novel concept of the edge cloud has recently been introduced to reduce transmission costs in mobile cloud computing services. Heterogeneous networks with diverse radio access networks will be pervasive in the future. In this paper, we propose a spatial and temporal computation offloading decision algorithm (ST-CODA) in edge cloud-enabled heterogeneous networks. In ST-CODA, a mobile device decides where and when to process tasks by means of a Markov decision process with the consideration of the processing time and energy consumption of different computation nodes and the transmission cost in heterogeneous networks. Extensive evaluation results are given to demonstrate the effectiveness of the ST-CODA in terms of the transmission cost, the energy efficiency of the mobile device, and the number of tasks that can be processed before their deadline.

Lightweight Opportunistic Mobile Data Offloading

Current cellular networks are overloaded due to the increasing number of smartphones and demands for bandwidth-eager multimedia content. Upgrading the existing infrastructure of the cellular system is the most straight forward solution to meet the growing demand. Apart from this, offloading mobile data through Wi-Fi can be a feasible solution. Mobile offloading via Wi-Fi is the latest emerging trend in research and industry. In this article, the authors have proposed a framework for mobile data offloading for both cellular and Wi-Fi networks. The authors have introduced a daemon process-based approach to make the entire process lightweight by using a suitable offloading decision algorithm. This article then formulates a mathematical model to evaluate its feasibility and accuracy for achieving optimum performance.

MCloud: A Context-Aware Offloading Framework for Heterogeneous Mobile Cloud

Mobile cloud computing (MCC) has become a significant paradigm for bringing the benefits of cloud computing to mobile devices’ proximity. Service availability along with performance enhancement and energy efficiency are primary targets in MCC. This paper proposes a code offloading framework, called mCloud, which consists of mobile devices, nearby cloudlets and public cloud services, to improve the performance and availability of the MCC services. The effect of the mobile device context (e.g., network conditions) on offloading decisions is studied by proposing a context-aware offloading decision algorithm aiming to provide code offloading decisions at runtime on selecting wireless medium and appropriate cloud resources for offloading. We also investigate failure detection and recovery policies for our mCloud system. We explain in details the design and implementation of the mCloud prototype framework. We conduct real experiments on the implemented system to evaluate the performance of the algorithm. Results indicate the system and embedded decision algorithm are able to provide decisions on selecting wireless medium and cloud resources based on different context of the mobile devices, and achieve significant reduction on makespan and energy, with the improved service availability when compared with existing offloading schemes.

Multiuser Joint Task Offloading and Resource Optimization in Proximate Clouds

Proximate cloud computing enables computationally intensive applications on mobile devices, providing a rich user experience. However, remote resource bottlenecks limit the scalability of offloading, requiring optimization of the offloading decision and resource utilization. To this end, in this paper, we leverage the variability in capabilities of mobile devices and user preferences. Our system utility metric is a measure of quality of experience (QoE) based on task completion time and energy consumption of a mobile device. We propose a heuristic offloading decision algorithm (HODA), which is semidistributed and jointly optimizes the offloading decision, and communication and computation resources to maximize system utility. Our main contribution is to reduce the problem to a submodular maximization problem and prove its NP-hardness by decomposing it into two subproblems: 1) optimization of communication and computation resources solved by quasiconvex and convex optimization and 2) offloading decision solved by submodular set function optimization. HODA reduces the complexity of finding the local optimum to $O(K^{3})$ , where $K$ is the number of mobile users. Simulation results show that HODA performs within 5% of the optimal on average. Compared with other solutions, HODA's performance is significantly superior as the number of users increases.

An adaptive mobile cloud computing framework using a call graph based model

The use of mobile applications and their functionality are increasing day by day but mobile devices are still inferior to ordinary computers in terms of memory and processor capacity. Furthermore, the rapid depletion of the mobile devices’ energy is still a major problem. Performance and energy shortcomings of mobile devices can be improved by using surrogate or cloud computing technologies. In particular, computation and memory intensive real time applications would be efficiently run by utilizing the resources of a remote server. In this paper, a novel offloading framework based on the Inversion of Control mechanism is developed to overcome the shortcomings and limitations of the current offloading approaches published in the literature. The proposed offloading framework reduces the burden on programmers. It implements application partitioning and code offloading via remote proxy classes and seamlessly provides callback functionality. In an application, it is possible to migrate different combinations of application components to remote servers. Some of these combinations can be productive and others can be counterproductive for offloading. In order to decide on components to be offloaded, a call graph model based on the collaboration of application’s classes is developed. An offloading decision algorithm is presented to determine the classes to be offloaded by finding the best application partitioning in the graph. The framework and the graph model are evaluated by several experiments. Experimental results show that the proposed graph model fits well to the application partitioning problem. It has also been shown that offloading the optimal combination of components to remote servers can considerably reduce the execution time and energy consumption of mobile devices.