/pdf/ieee-wireless-communications-and-networking-conference-2fkoe7n959.pdf

IEEE Wireless Communications and Networking Conference

With the proliferation of the Internet of Things (IoT) and the wide penetration of wireless networks, the surging demand for data communications and computing calls for the emerging edge computing paradigm. By moving the services and functions located in the cloud to the proximity of users, edge computing can provide powerful communication, storage, networking, and communication capacity. The resource scheduling in edge computing, which is the key to the success of edge computing systems, has attracted increasing research interests. In this paper, we survey the state-of-the-art research findings to know the research progress in this field. Specifically, we present the architecture of edge computing, under which different collaborative manners for resource scheduling are discussed. Particularly, we introduce a unified model before summarizing the current works on resource scheduling from three research issues, including computation offloading, resource allocation, and resource provisioning. Based on two modes of operation, i.e., centralized and distributed modes, different techniques for resource scheduling are discussed and compared. Also, we summarize the main performance indicators based on the surveyed literature. To shed light on the significance of resource scheduling in real-world scenarios, we discuss several typical application scenarios involved in the research of resource scheduling in edge computing. Finally, we highlight some open research challenges yet to be addressed and outline several open issues as the future research direction.

Resource Scheduling in Edge Computing: A Survey

Achieving long battery lives or even self sustainability has been a long standing challenge for designing mobile devices. This paper presents a novel solution that seamlessly integrates two technologies, mobile cloud computing and microwave power transfer (MPT), to enable computation in passive low-complexity devices such as sensors and wearable computing devices. Specifically, considering a single-user system, a base station (BS) either transfers power to or offloads computation from a mobile to the cloud; the mobile uses harvested energy to compute given data either locally or by offloading. A framework for energy efficient computing is proposed that comprises a set of policies for controlling CPU cycles for the mode of local computing, time division between MPT and offloading for the other mode of offloading, and mode selection. Given the CPU-cycle statistics information and channel state information (CSI), the policies aim at maximizing the probability of successfully computing given data, called computing probability, under the energy harvesting and deadline constraints. The policy optimization is translated into the equivalent problems of minimizing the mobile energy consumption for local computing and maximizing the mobile energy savings for offloading which are solved using convex optimization theory. The structures of the resultant policies are characterized in closed form. Furthermore, given non-causal CSI, the said analytical framework is further developed to support computation load allocation over multiple channel realizations, which further increases computing probability. Last, simulation demonstrates the feasibility of wirelessly powered mobile cloud computing and the gain of its optimal control.

Energy Efficient Mobile Cloud Computing Powered by Wireless Energy Transfer

Edge computing (EC) has recently emerged as a promising paradigm that supports resource-hungry Internet of Things (IoT) applications with low latency services at the network edge. However, the limited capacity of computing resources at the edge server poses great challenges for scheduling application tasks. In this paper, a task scheduling problem is studied in the EC scenario, and multiple tasks are scheduled to virtual machines (VMs) configured at the edge server by maximizing the long-term task satisfaction degree (LTSD). The problem is formulated as a Markov decision process (MDP) for which the state, action, state transition, and reward are designed. We leverage deep reinforcement learning (DRL) to solve both time scheduling (i.e., the task execution order) and resource allocation (i.e., which VM the task is assigned to), considering the diversity of the tasks and the heterogeneity of available resources. A policy-based REINFORCE algorithm is proposed for the task scheduling problem, and a fully-connected neural network (FCN) is utilized to extract the features. Simulation results show that the proposed DRL-based task scheduling algorithm outperforms the existing methods in the literature in terms of the average task satisfaction degree and success ratio.

https://www.mdpi.com/1424-8220/21/5/1666/pdf

Deep Reinforcement Learning-Based Task Scheduling in IoT Edge Computing

In today's world, the internet of things (IoT) is developing rapidly. Wireless sensor network (WSN) as an infrastructure of IoT has limitations in the processing power, storage, and delay for data transfer to cloud. The large volume of generated data and their transmission between WSNs and cloud are serious challenges. Fog computing (FC) as an extension of cloud to the edge of the network reduces latency and traffic; thus, it is very useful in IoT applications such as healthcare applications, wearables, intelligent transportation systems, and smart cities. Resource allocation and task scheduling are the NP-hard issues in FC. Each application includes several modules that require resources to run. Fog devices (FDs) have the ability to run resource management algorithms because of their proximity to sensors and cloud as well as the proper processing power. In this paper, we review the scheduling strategies and parameters as well as providing a greedy knapsack-based scheduling (GKS) algorithm for allocating resources appropriately to modules in fog network. Our proposed method was simulated in iFogsim as a standard simulator for FC. The results show that the energy consumption, execution cost, and sensor lifetime in GKS are better than those of the first-come-first-served (FCFS), concurrent, and delay-priority algorithms.

/pdf/low-latency-and-energy-efficient-scheduling-in-fog-based-iot-493rq3jd61.pdf

Jianhang Tang

Papers

Joint optimization of data placement and scheduling for improving user experience in edge computing

Collaborative cache allocation and task scheduling for data-intensive applications in edge computing environment

Load balance based workflow job scheduling algorithm in distributed cloud

Dynamic resource allocation strategy for latency-critical and computation-intensive applications in cloud–edge environment

Radio and computing resource allocation with energy harvesting devices in mobile edge computing environment