Multi-Objective Firefly Simulated Annealing (MOFSA) Based Inter-Data Center Network Dimensioning Under Cloud Computing Environment

Author Name(s): *K. Srilakshmi, P. Bhargavi
Author Email: srilakshmi21kakarla@gmail.com

Abstract

Cloud Computing (CC) presents a big chance for Inter-Data Center (IDC) network to modernize its infrastructure, systems and services. In addition these IDC network, the allotment and liberate of resources by various cloud customers bring about noteworthy energy utilization, less security and Virtual Machine (VM) position of at the IDC that consecutively enhances the Operational expenditures (Opex) of the operators. In order to resolve these troubles, this research merges energy effectiveness, virtual machine assignment, immigration restraints and Time of Use (ToU) awareness, and presents a new virtualization method, that is to say Time of Use Resource Provisioning (ToURP) awareness ToUP for an IDC network on top of an Internet Protocol (IP) on Wavelength Division Multiplexing (WDM) backbone. At first, we propose a Mixed Integer Linear Programming (MILP) formulation to model the best possible performance of ToURP. On the other hand additionally the traffic jammed between two backbone nodes, upstream user insists predestined to data centers and downstream data center insists beginning from numerous data centers. As the IDC network requires to be rearranged in polynomial time, we present a Multi-Objective Firefly Simulated Annealing (MOFSA) based algorithm for depiction of memory, storage, and communication resources in IDC network. This holistic technique is to sustainable ToURP, the toolkit is intended to give a base for a consistent and hopeful CC. They assess ToURP under diverse situations, and mathematical outcomes prove that noteworthy Opex savings could be attained whilst insists could be stipulated with less energy utilization in the IDC.

Keywords

Cloud Computing (CC), Time of Use Resource Provisioning (ToURP), Service Level Agreement (SLA), Mixed Integer Linear PROGRAMMING (MILP), Inter-Data Center (IDC) Network, Multi-objective Firefly Simulated Annealing (MOFSA).

Introduction

Cloud Computing (CC) is identified as presently one among the very trendy techniques existing; customers make use of the idea of “pay-as-you-go” for applications, computing and storage resources [1]. The ecological effect of data centers is important and the development forces of the Information Technology (IT) sector recommends professionals to deal with this issue. Since the quantity of data demanding applications amplify in the Cloud, use of huge amounts of physical servers boost radically not only for computing purpose but for as well replication intentions [2].

Conclusion

In this research, we analyzed the Time-of-Use resource provisioning (ToRUP) – aware design of a virtual IDC network to reduce the electricity bills of the operators and decrease the energy utilization all through the CC Environment. On the other hand the ToURP awareness based Multi-Objective Firefly Simulated Annealing (MOFSA) based meta-heuristic algorithm that intends to get a DC workload sharing map. Primarily, we propose a MILP formulation to form the best possible conduct of ToURP. On the other hand additionally the traffic amid two backbone nodes, upstream user demands intended to data centers and downstream data center insists beginning from a lot of data centers. Power consumption value of the proposed ToRUP is 4871 kW for estimation time of 18 hrs , whereas the power consumption value of other ToUP and MILP methods are 5421 kW and 5935 kW respectively. Furthermore, we have proved that ToRUP brings about additional power utilization savings in the Cloud while every data center is forced to maintain some quantity of its original workload. The upcoming work encloses modeling and designing VM immigration procedures, developing new task and VM scheduling techniques for the model, and enhancements of the models implementation.

836 total views, no views today

Download PDF File

About the author: admin