Task Priority-Based Cached-Data Prefetching and Eviction Mechanisms for Performance Optimization of Edge Computing Clusters

The rapid evolution of the Internet of Things (IoT) and the development of cloud computing have endorsed a new computing paradigm called edge computing, which brings the computing resources to the edge of the network. Due to low computing power and small data storage at the edge nodes, the task must be assigned to the computing nodes, where their associated data is available, to reduce overheads caused by data transmissions in the network. The proposed scheme named task priority-based data-prefetching scheduler (TPDS) tries to improve the data locality through available cached and prefetching data for offloading tasks to the edge computing nodes. The proposed TPDS prioritizes the tasks in the queue based on the available cached data in the edge computing nodes. Consequently, it increases the utilization of cached data and reduces the overhead caused by data eviction. The simulation results show that the proposed TPDS can be effective in terms of task scheduling and data locality.

Download Full-text

Microservice Security Agent Based On API Gateway in Edge Computing

Sensors ◽

10.3390/s19224905 ◽

2019 ◽

Vol 19 (22) ◽

pp. 4905 ◽

Cited By ~ 3

Author(s):

Rongxu Xu ◽

Wenquan Jin ◽

Dohyeun Kim

Keyword(s):

Data Storage ◽

User Authentication ◽

Edge Computing ◽

Security And Privacy ◽

Agent Based ◽

Computing Platform ◽

Authentication Mechanism ◽

Computing Paradigm ◽

Security Agent ◽

Secure Authentication

Internet of Things (IoT) devices are embedded with software, electronics, and sensors, and feature connectivity with constrained resources. They require the edge computing paradigm, with modular characteristics relying on microservices, to provide an extensible and lightweight computing framework at the edge of the network. Edge computing can relieve the burden of centralized cloud computing by performing certain operations, such as data storage and task computation, at the edge of the network. Despite the benefits of edge computing, it can lead to many challenges in terms of security and privacy issues. Thus, services that protect privacy and secure data are essential functions in edge computing. For example, the end user’s ownership and privacy information and control are separated, which can easily lead to data leakage, unauthorized data manipulation, and other data security concerns. Thus, the confidentiality and integrity of the data cannot be guaranteed and, so, more secure authentication and access mechanisms are required to ensure that the microservices are exposed only to authorized users. In this paper, we propose a microservice security agent to integrate the edge computing platform with the API gateway technology for presenting a secure authentication mechanism. The aim of this platform is to afford edge computing clients a practical application which provides user authentication and allows JSON Web Token (JWT)-based secure access to the services of edge computing. To integrate the edge computing platform with the API gateway, we implement a microservice security agent based on the open-source Kong in the EdgeX Foundry framework. Also to provide an easy-to-use approach with Kong, we implement REST APIs for generating new consumers, registering services, configuring access controls. Finally, the usability of the proposed approach is demonstrated by evaluating the round trip time (RTT). The results demonstrate the efficiency of the system and its suitability for real-world applications.

Download Full-text

Analysis of Security Issues of Edge Computing Based Internet of Things Applications

10.31219/osf.io/8wzha ◽

2020 ◽

Author(s):

Shamim Muhammad ◽

Inderveer Chana ◽

Supriya Thilakanathan

Keyword(s):

Internet Of Things ◽

Edge Computing ◽

The Internet ◽

Security Vulnerabilities ◽

Computing Power ◽

Security Issues ◽

Computing Paradigm ◽

Improved Performance ◽

The Internet Of Things ◽

Common Security

Edge computing is a technology that allows resources to be processed or executed close to the edge of the internet. The interconnected network of devices in the Internet of Things has led to an increased amount of data, increasing internet traffic usage every year. Also, edge computing is driving applications and computing power away from the integrated points to areas close to users, leading to improved performance of the application. Despite the explosive growth of the edge computing paradigm, there are common security vulnerabilities associated with the Internet of Things applications. This paper will evaluate and analyze some of the most common security issues that pose a serious threat to the edge computing paradigm.

Download Full-text

Edge Architecture Integration of Technologies

Cases on Edge Computing and Analytics - Advances in Computational Intelligence and Robotics ◽

10.4018/978-1-7998-4873-8.ch001 ◽

2021 ◽

pp. 1-30

Author(s):

Sandhya Devi R. S. ◽

Vijaykumar V. R. ◽

Sivakumar P. ◽

Neeraja Lakshmi A. ◽

Vinoth Kumar B.

Keyword(s):

Data Privacy ◽

Response Times ◽

Management Strategies ◽

Cost Savings ◽

Edge Computing ◽

Battery Life ◽

Computing Platform ◽

Computing Paradigm ◽

Time Latency ◽

The Internet Of Things

The enormous growth of the internet of things (IoT) and cloud-based services have paved the way for edge computing, the new computing paradigm which processes the data at the edge of the network. Edge computing resolves issues related to response time, latency, battery life limitation, cost savings for bandwidth, as well as data privacy and protection. The architecture brings devices and data back to the consumer. This model of computing as a distributed IT system aims at satisfying end-user demands with faster response times by storing data closer to it. The enormous increase in individuals and locations, connected devices such as appliances, laptops, smartphones, and transport networks that communicate with each other has raised exponentially. Considering these factors in this chapter, edge computing architecture along with the various components that constitute the computing platform are discussed. The chapter also discusses resource management strategies deliberate for edge computing devices and integration of various computing technologies to support efficient IoT architecture.

Download Full-text

Dependable and Secure Voting Mechanism in Edge Computing

Future Internet ◽

10.3390/fi11120262 ◽

2019 ◽

Vol 11 (12) ◽

pp. 262

Author(s):

Pedro A.R.S. Costa ◽

Marko Beko

Keyword(s):

Middle Layer ◽

Edge Computing ◽

Cloud Environment ◽

Computing Paradigm ◽

Edge Network ◽

And Storage ◽

The Internet Of Things ◽

Multi Cloud ◽

Voting Mechanism ◽

Open Issues

Edge computing is a distributed computing paradigm that encompasses data computing and storage and is performed close to the user, efficiently guaranteeing faster response time. This paradigm plays a pivotal role in the world of the Internet of Things (IoT). Moreover, the concept of the distributed edge cloud raises several interesting open issues, e.g., failure recovery and security. In this paper, we propose a system composed of edge nodes and multiple cloud instances, as well as a voting mechanism. The multi-cloud environment aims to perform centralized computations, and edge nodes behave as a middle layer between edge devices and the cloud. Moreover, we present a voting mechanism that leverages the edge network to validate the performed computation that occurred in the centralized environment.

Download Full-text