Network reliability evaluation for a distributed network with edge computing

Abstract

Reliability and speed are crucial for ensuring robustness in a computer network, which is used for sharing information and resources within a business. The two most recent and important developments in the field of computing are cloud computing and edge computing. In this study, a modern distributed network comprising computers, internet of things (IoT), edge servers, and cloud servers for data transmission was investigated. A distributed network can be modeled as a graph with nodes and arcs, in which each arc represents a transmission line, and each node represents an IoT device, an edge server, or a cloud server. The data are transmitted through these transmission lines which are combined with numerous physical lines such as fiber cables. Therefore, each transmission line has several possible capacities and the capacity of each arc in the distributed network can be regarded as a multistate. Such the distributed network is called a multi-state distributed network (MDN). In an MDN, the data produced by IoT devices are transmitted to the edge servers. After being processed and compressed by the edge servers, the data are then transmitted to the cloud servers for advance computing. The purpose of this study is to first construct a model of a network in order to elucidate the flow relationship among the IoT devices, edge servers, and cloud servers and subsequently develop an algorithm to evaluate network reliability, which is the probability of the data being successfully processed by the MDN.