An Efficient Cloud Data Center Allocation to the Source of Requests

Introduction

Cloud data centers are the main source of variety of services ranging from computational to network and are delivered as on-demand services to users. The requests for these cloud services can come from different parts of the world (Rawal et al., 2011 and Rawal et al., 2013). The term source of requests/clients denote the users who make requests to various cloud data center services (Shen et al., 2017; Shen et al., 2016). The distance between the cloud data center and the source of requests is a major factor influencing the quality of service in terms of response time and latency. Cloud data center allocation is one of the major issues in cloud computing. An efficient allocation of cloud data center to the source of requests may improve the quality of services. However, there have been only few approaches that consider the cloud data center allocation to the source of requests.

In recent literature, Joseph Doyle et al. (2013) has proposed the source of requests assignment to the closest cloud data center to reduce the carbon emission, but they modeled cloud data center as a complete graph, which is unrealistic. They modelled both the networking and computational components of the infrastructure as a graph and proposed a system which utilizes Voronoi partitions to determine how source requests to be routed to appropriate data center based on the relative priorities of the cloud operator for latency purposes. This allows routing of the traffic to the data center that is closest in terms of geographical distance, costs the least in terms of power, and emits the smallest amount of carbon for a given request to lower carbon emissions and operational cost. This work examined the electricity cost, carbon emissions, and average service request time for a variety of scenarios.

(Judit Bar-Ilan et al. (1992), Randeep Bhatia et al. (1998), Reza Zanjirani Farahani et al. (2010), Irina Harris et al. (2014) provided solutions for facility location problems. They have considered distributing the clients to centers as balanced as possible, but they have overlooked the distance between clients and centers, which is also essential for faster service provisioning, hence there arise a need to develop an efficient method which allocates closest cloud data centers to the source of requests and keeps the load of the cloud data center as balanced as possible.

This paper models cloud data center as a graph and proposes an algorithm - modified Breadth First Search (MBFS) to efficiently allocate cloud data centers to the source of requests based on a cost threshold. Here the term cost refers to the distance between the cloud data center and the source of request. The aim is to allocate each source of request to a cloud data center in a faster manner based on cost threshold. The cost threshold is calculated as the average path length between cloud data centers and the source of requests of modified Voronoi approach. This may lead to faster service provisioning of the cloud data centers to the source of requests. The performance of proposed algorithm is compared with that of modified Voronoi and modified graph-based K-Means algorithms for various graph instances.

The following contributions are made in this paper:

• 1.

  Modified Breadth First Search algorithm is proposed;

• 2.

  Modified Voronoi algorithm is proposed;

• 3.

  Modified graph-based K-Means algorithm is proposed;

• 4.

  A random graph generator is constructed;

• 5.

  Comparison between approaches 1, 2 and 3 is done in terms of average time taken for allocation, average cost of cloud data centers and average load of cloud data centers.