Network Aware Virtual Machine Migration by PSO Optimization
Journal Title: International Journal for Research in Applied Science and Engineering Technology (IJRASET) - Year 2017, Vol 5, Issue 5
Abstract
A good VM migration algorithm can greatly improve network performance and scalability. Only a few studies presently focus on the network-aware VM migration (NetVMM) problem. The NetVMM problem is a type of a multiple-knapsack problem. Thus, finding an optimal solution in polynomial time is not practical. Our goal is to find an approximation solution to this NPcomplete problem which is energy efficient. In this era of technology, some energy efficient techniques are needed. Hence we were motivated to carry out this project and work for energy efficient environment. Our objective to find is an approximate optimal solution which is energy efficient through repeated iterations to make it a good solution for the VM migration problem. We will make a Matlab program for showing the output of the PSO optimization algorithms used. In computing and research, a genetic algorithm (GA) may be a meta-heuristic galvanized by the method of natural action that belongs to the larger category of organic process algorithms. Genetic algorithms are unremarkably accustomed generate high-quality solutions to improvement and search issues by hoping on bio-inspired operators like mutation, crossover and choice. In a genetic algorithm, a population of candidate solutions to an optimization problem evolves toward better solutions. Each candidate solution has a set of properties. Particle swarm optimization (PSO) solves a optimization scenario by having a number of candidate solutions, here virtual machines, and migrating these virtual machines around in the search-space according to some derived formulae over the virtual machines’ position and velocity. We will be implementing the PSO optimization algorithm in Matlab environment to solve the problem of cost optimization of the data centers, so that optimal number of tasks are divided in each virtual machine, hence balancing the cost and the network.
Authors and Affiliations
Sarthak Tandon, Chandan Kesarwani, Paridhi Srivastava, Abhishek Suryan, Swathi J N
Keywords
Related Articles
- EP ID EP24122
- DOI -
- Views 295
- Downloads 8
Finite State Machines
Finite state machines (FSMs) are a common presence in digital circuit design. However, they can be very useful also for the software developer. Actual operating systems and application software are event-based and commu...
Improved Chord Algorithm
A peer-to-peer networks (P2P) has increased popularity and is now more than a day. Peer-to-peer overlay network on top of the system Device or abstract physical network topology. The network consists of two types of str...
High Efficiency Intelligent Street Lighting System Using a Zigbee Network and GSM
The proposed street lighting system is highly energy efficient and automated. It uses ZigBee-based wireless devices due to which street lamp system becomes more efficient. ZigBee network is connected to GSM through comp...
FEA Study of Pre-Fill Valve
Pre-fill valve is Pilot-operated check valve which is designed to permit free flow in one direction and to block return flow, unless pilot pressure is applied. However, under pilot pressure, flow is permitted in both di...
Thermal Analysis of Exhaust Heat Exchanger in Automobile Thermo Electric Generator
By the view of increasing fuel costs and environmental issues, it is a requirement to manage the power utilization. In here it is mere focus on the automobile that, they take major contribution out of all power produced...