Communication-Aware Traffic Stream Optimization for Virtual Machine Placement in Cloud Datacenters with VL2 Topology

Subject Areas : B.3. Communication/Networking and Information Technology

Sara  Farzai ¹ (Department of Computer Engineering, Sari Branch, Islamic Azad University, Sari, Iran)
Mirsaeid  Hosseini Shirvani ² (Department of Computer Engineering, Sari Branch, Islamic Azad University, Sari, Iran)
Mohsen  Rabbani ³ (Department of Mathematics, Sari Branch, Islamic Azad university, Sari, Iran)

Received: 2020-09-05 Accepted : 2020-09-21 Published : 2020-08-01

Keywords: cloud computing, Meta-Heuristic Algorithms, network traffic management, virtual machine placement, VL2,

Abstract :

By pervasiveness of cloud computing, a colossal amount of applications from gigantic organizations increasingly tend to rely on cloud services. These demands caused a great number of applications in form of couple of virtual machines (VMs) requests to be executed on data centers’ servers. Some of applications are as big as not possible to be processed upon a single VM. Also, there exists several distributed applications such as MapReduce projects which exploit much number of VMs dispersed over physical machines (PMs) attached with high speed networks. These types of VMs involve mutual traffic transferring which is completely processed as an atomic application. High volume of traffic transfer among VMs may saturate network links and leads performance bottleneck for both data center and applications which seriously threat users’ service level agreement (SLA). Furthermore, communication energy consumption increases when network devices are heavily in use. This paper addresses the virtual machine placement (VMP) problem by considering inter-VM communications on VL2 topology. This is an optimization problem with the aim of network traffic transferring minimization. Dependent VMs are tried to be co-hosted or to be placed in close neighborhoods to minimize the amount of total traffic streaming over the network. A combined meta-heuristic approach based and ACO and GA algorithms is employed to solve the problem. The results of simulations imply the superiority of our proposed approach in comparison with other state-of-the-art approaches in terms of reducing total traffic flow, saving energy, and declining resource dissipation in servers.