Informatics Publishing Limited

T. Lavanya Suja ¹, V. Savithri ²

Affiliations
1 Mother Teresa Women's University, Kodaikanal - 624102, Tamil Nadu, IN
2 Women's Christian College, Nungambakkam, Chennai - 600006, Tamil Nadu, IN

Abstract

Objectives: The aim of the paper is to propose a novel approach for Virtual Cluster (VC) migration in cloud computing which achieves better performance in less time and incorporating all Service Level Agreements. Method: Cloud Computing (CC) is rapidly gaining momentum and has spread its flavors and dimensions in wide discipline of Information Technology (IT) industry. When a new technology is spreading its wings to various places the functionalities realized by it throws challenges too. In an attempt to tap the whole potential of CC these challenges are addresses by professionals from IT industry and academia. Out of the myriad number of challenges, one is the need for migration of Virtual Clusters (VC) from one physical machine to another physical machine in spite of fault tolerance, load balancing, complying with Service Level Agreements (SLA). After studying and analyzing the various existing approaches for VC migration it is proposed that Backtracking algorithm outperforms the existing ones in terms of time and performance. Findings: The implementation of Backtracking algorithm proves the fact that the above objectives are achieved and the test results through line graphs below clarifies the fact further. The novelty of the algorithm and the compact lines of coding achieve the optimal solution for VC migration with satisfying SLA and time constraints. The comparison chart on feature analysis clearly list down the additional characteristics of the backtracking algorithm which adds value and uniqueness to the optimal solution. The flowchart aids to the development of algorithm and its easy implementation. The test data is run for various memory sizes of VC and VM therefore checked for all type of situations arising at real time. All the solutions are optimal in case of time and efficiency when compared with other exciting techniques like serial and greedy approaches. Application/Improvement: Though a handful of approaches are there in CC field for VC migration, not all techniques provides a balance between performance and time while the proposed backtracking algorithm does it with a unique solution. Also it is one of the few algorithms which satisfy all the SLA constraints and one of a kind to accommodate any number of SLA into the algorithm.

Keywords

Backtracking Algorithm, Cloud Computing, PM, VC, Virtual Cluster, VC Migration, VM, Virtual Machines.

Full Text

   

V. Savithri ¹, S. Purushothaman ²

Affiliations
1 Mother Teresa Women's University, Kodaikanal, IN
2 Udaya School of Engineering, Kanyakumari District-629204, IN

Abstract

The objective of this work is to develop and implement an investigation carried out on carotid artery. The proposed method categorizes the carotid artery subjects into normal and diseased subjects. Ultrasound image videos of the artery are used as the data. The frames of the video are processed to know the properties of the artery. In order to extract properties, image processing techniques have been used on each frame. The features extracted from the frames are consolidating to know the conditions of the artery. These features are accurately analysed to know the status of the artery by using intelligent techniques like artificial neural network and fuzzy logic. In artificial neural network using back propagation algorithm and by fuzzy logic system provides higher classification efficiency with minimum training and testing time. It helps in developing medical decision system for ultrasound artery images. It can also be used as secondary observer in clinical decision making.

Keywords

Artery, Boundary Detection, Intima Media Thickness, Neural Network, Ultrasonic.

Full Text