An Efficient Edge Server Deployment Technique In Edge Computing Using Graph Theory

Abstract

Cloud computing is a paradigm used to store and share the data in the data centers to the users through
the Internet. It provides many services to the users like IaaS(Infrastructure as a Service), PaaS (Platform
as a Service), SaaS (Software as a Service) etc , but distributing the centralized data over the world is a
complex process, it takes insufficient time. To overcome this, a new paradigm Edge Computing was
introduced. Here the data are processed in the edge of the network near the source of the data, so data
processing speed is increased. In Edge computing there was no standard architecture followed but some
accepted architectures were allowed to process the data. This paper focuses the three layered architecture
(Edge nodes, Edge servers, clouds are the main components of this architecture). This paper proposes the
placement of new Edge server in the distributed clustered Edge data servers in the middle layer. For
placethenew Edge server, here Dijkstra Algorithm is used to find the minimum distance in the architecture
and then place the new server. Placing the Edge server in that particular location will improve the
performance of the distributed Edge servers, since data travelling distance is reduced and losing the data
is also reduced to minimum level. Dijkstra is a famous graph theory algorithm which is mainly used to find
the minimum distance between the nodes in graph theory concept. Its performance also analyzed with
Floyd-Warshall algorithm. Hope this paper leads to some other techniques to find the minimum distance
to place the edge server.