Topic
Edge computing
About: Edge computing is a research topic. Over the lifetime, 11657 publications have been published within this topic receiving 148533 citations.
Papers published on a yearly basis
Papers
More filters
01 Oct 2018
TL;DR: KubeEdge infrastructure connects and coordinates two computing environments for applications leveraging both computing resources to achieve better performance and user experience, and provides the network protocol infrastructure and the same runtime environment on the edge as in the cloud.
Abstract: In this paper, we introduce an infrastructure in edge computing environment, KubeEdge, to extend cloud capabilities to the edge. In the new form of cloud architecture, Cloud consists of computing resources both at centralized data centers and at distributed edges. KubeEdge infrastructure connects and coordinates two computing environments for applications leveraging both computing resources to achieve better performance and user experience. Technically, KubeEdge provides the network protocol infrastructure and the same runtime environment on the edge as in the cloud, which allows the seamless communication of applications with components running on edge nodes as well as cloud servers. It also allows the existing cloud services and cloud development model to be adopted at edge. Based on Kubernetes [1], KubeEdge architecture includes a network protocol stack called KubeBus, a distributed metadata store and synchronization service, and a lightweight agent (EdgeCore) for the edge. KubeBus is designed to have its own implementation of OSI network protocol layers, which connects servers at edge and VMs in the cloud as one virtual network. KubeBus provides a unified multitenant communication infrastructure with fault tolerance and high availability. The distributed metadata store and sync service is designed to support the offline scenario when edge nodes are not connected to the cloud. EdgeController component in KubeEdge architecture is a controller plugin for Kubernetes [1] to manage remote edge nodes and cloud VMs as one logical cluster, which enables KubeEdge to schedule, deploy and manage container applications across edge and cloud with the same API.
121 citations
TL;DR: This paper proposes an approach that involves device-driven and human-driven intelligence as key enablers to reduce energy consumption and latency in fog computing via two case studies that make use of the machine learning to detect user behaviors and perform adaptive low-latency Medium Access Control (MAC)-layer scheduling among sensor devices.
121 citations
08 Jul 2019
TL;DR: This paper introduces the main technologies supporting the Edge paradigm, survey existing issues, introduce relevant scenarios, and discusses benefits and caveats of the different existing solutions in the above introduced scenarios.
Abstract: Edge and Fog Computing will be increasingly pervasive in the years to come due to the benefits they bring in many specific use-case scenarios over traditional Cloud Computing. Nevertheless, the security concerns Fog and Edge Computing bring in have not been fully considered and addressed so far, especially when considering the underlying technologies (e.g. virtualization) instrumental to reap the benefits of the adoption of the Edge paradigm. In particular, these virtualization technologies (i.e. Containers, Real Time Operating Systems, and Unikernels), are far from being adequately resilient and secure. Aiming at shedding some light on current technology limitations, and providing hints on future research security issues and technology development, in this paper we introduce the main technologies supporting the Edge paradigm, survey existing issues, introduce relevant scenarios, and discusses benefits and caveats of the different existing solutions in the above introduced scenarios. Finally, we provide a discussion on the current security issues in the introduced context, and strive to outline future research directions in both security and technology development in a number of Edge/Fog scenarios.
121 citations
Posted Content•
TL;DR: The role of edge computing in realizing the vision of smart cities is highlighted, and several indispensable open challenges along with their causes and guidelines are discussed, serving as future research directions.
Abstract: Recent years have disclosed a remarkable proliferation of compute-intensive applications in smart cities. Such applications continuously generate enormous amounts of data which demand strict latency-aware computational processing capabilities. Although edge computing is an appealing technology to compensate for stringent latency related issues, its deployment engenders new challenges. In this survey, we highlight the role of edge computing in realizing the vision of smart cities. First, we analyze the evolution of edge computing paradigms. Subsequently, we critically review the state-of-the-art literature focusing on edge computing applications in smart cities. Later, we categorize and classify the literature by devising a comprehensive and meticulous taxonomy. Furthermore, we identify and discuss key requirements, and enumerate recently reported synergies of edge computing enabled smart cities. Finally, several indispensable open challenges along with their causes and guidelines are discussed, serving as future research directions.
120 citations
TL;DR: A FANN-on-MCU, an open-source toolkit built upon the fast artificial neural network (FANN) library to run lightweight and energy-efficient neural networks on microcontrollers based on both the ARM Cortex-M series and the novel RISC-V-based parallel ultralow-power (PULP) platform is presented.
Abstract: The growing number of low-power smart devices in the Internet of Things is coupled with the concept of “edge computing” that is moving some of the intelligence, especially machine learning, toward the edge of the network. Enabling machine learning algorithms to run on resource-constrained hardware, typically on low-power smart devices, is challenging in terms of hardware (optimized and energy-efficient integrated circuits), algorithmic, and firmware implementations. This article presents a FANN-on-MCU, an open-source toolkit built upon the fast artificial neural network (FANN) library to run lightweight and energy-efficient neural networks on microcontrollers based on both the ARM Cortex-M series and the novel RISC-V-based parallel ultralow-power (PULP) platform. The toolkit takes multilayer perceptrons trained with FANN and generates code targeted to low-power microcontrollers. This article also presents detailed analyses of energy efficiency across the different cores, and the optimizations to handle different network sizes. Moreover, it provides a detailed analysis of parallel speedups and degradations due to parallelization overhead and memory transfers. Further evaluations include experimental results for three different applications using a self-sustainable wearable multisensor bracelet. The experimental results show a measured latency in the order of only a few microseconds and power consumption of a few milliwatts while keeping the memory requirements below the limitations of the targeted microcontrollers. In particular, the parallel implementation on the octa-core RISC-V platform reaches a speedup of $22\times $ and a 69% reduction in energy consumption with respect to a single-core implementation on Cortex-M4 for continuous real-time classification.
120 citations