Informatics Publishing Limited

A. Suthan ¹, D. Kesavaraja ²

Affiliations
1 Department of Computer Applications, DrSivanthi Aditanar College of Engineering , Tiruchendur–628215, IN
2 Department of Computer Science and Engineering, Dr Sivanthi Aditanar College of Engineering, Tiruchendur–628215, IN

Abstract

In the modern world Pervasive computing plays a vital role. Distributed computing finds its way in nearly every day to day operations of an individual. We have tried to produce a pervasive computing environment that provides accurate and continual service with the use of a distributed dependency graph generator. In order to provide an accurate service the faulty nodes are segregated from the main stream. In order to provide continual service the distributed dependency graph finds various surrogate nodes that are capable of performing the same functions. This is achieved by making the failed node to move to a safe state and perform sensing but not participate in the Decision Dependency Graph .Once the failed node has recovered; it is permitted to participate in the Decision Dependency Graph.

Keywords

Decision Dependency Graph, Fault Tolerant, Pervasive Computing, Surrogate Nodes.

Full Text

     

D. Kesavaraja ¹, A. Suthan ², J. Mark Jain ¹

Affiliations
1 Department of Computer Science and Engineering, Dr. Sivanthi Aditanar College of Engineering, Tiruchendur, Tamilnadu, IN
2 Department of Computer Applications, Dr. Sivanthi Aditanar College of Engineering, Tiruchendur, Tamilnadu, IN

Abstract

Congestion control has been always a major problem when network and data transmission gets involved. With the emergence of cloud computing the same problem applies to cloud computing environment too. In this paper we consider the problem of congestion control in cloud web computing environment which support both multirate multicast sessions and unicast sessions. Our work proposes an efficient congestion control mechanism using genetic algorithm, which identifies service state information of the communication. A Subserver justifies packet speed labeled by identifier and it decides the rapidness in which the receiver should receive the packet. Based on the receiver's requirement speed, the sub server splits the packets and forwards it. Also it identifies service state information for allocating bandwidth to the receivers. We propose an architecture that significantly reduces this implementation complexity but still achieves approximately fair bandwidth allocations. The congestion-control mechanism takes into account the diverse user requirements when different receivers within a multicast session have proper fitness functions in GA, but does not require the network to have any knowledge about the receiver utility functions.

Keywords

Cloud Computing, Congestion Control, Genetic Algorithm, Multi Cast, Traffic Flow, Packet Transfer.

Full Text

     

R. Shalin ¹, D. Kesavaraja ²

Affiliations
1 Department of Computer Science and Engineering, Dr. G.U. Pope College of Engineering, IN
2 Department of Computer Science and Engineering, Dr. Sivanthi Aditanar College of Engineering, IN

Abstract

The most preferred mode for communication of multimedia data is through the TCP/IP protocol. But on the other hand the TCP/IP protocol produces huge packet loss unavoidable due to network traffic and congestion. In order to provide a efficient communication it is necessary to recover the loss of packets. The proposed scheme implements Hash based FEC with auto XOR scheme for this purpose. The scheme is implemented through Forward error correction, MD5 and XOR for providing efficient transmission of multimedia data. The proposed scheme provides transmission high accuracy, throughput and low latency and loss.

Keywords

Encoding, FEC, Decoding, MD5, TCP/IP, XOR.

Full Text

     

D. Kesavaraja ¹, R. Balasubramanian ², R. S. Rajesh ², D. Sasireka ³

Affiliations
1 Department of Computer Science and Engineering, Dr. Sivanthi Aditanar College of Engineering, Tamil Nadu, IN
2 Department of Computer Science and Engineering, Manonmaniam Sundaranar University, Tamil Nadu, IN
3 Department of Information Technology, Jayamatha Engineering College, Tamil Nadu, IN

Abstract

This paper presents efficient and portable implementations of a useful image segmentation technique which makes use of the faster and a variant of the conventional connected components algorithm which we call parallel Components. In the Modern world majority of the doctors are need image segmentation as the service for various purposes and also they expect this system is run faster and secure. Usually Image segmentation Algorithms are not working faster. In spite of several ongoing researches in Conventional Segmentation and its Algorithms might not be able to run faster. So we propose a cluster computing environment for parallel image Segmentation to provide faster result. This paper is the real time implementation of Distributed Image Segmentation in Clustering of Nodes. We demonstrate the effectiveness and feasibility of our method on a set of Medical CT Scan Images. Our general framework is a single address space, distributed memory programming model. We use efficient techniques for distributing and coalescing data as well as efficient combinations of task and data parallelism. The image segmentation algorithm makes use of an efficient cluster process which uses a novel approach for parallel merging. Our experimental results are consistent with the theoretical analysis and practical results. It provides the faster execution time for segmentation, when compared with Conventional method. Our test data is different CT scan images from the Medical database. More efficient implementations of Image Segmentation will likely result in even faster execution times.

Keywords

Parallel Algorithms, Region Growing, Image Enhancement, Image Segmentation, Parallel Performance.

Full Text

     

A. Suthan ¹, D. Kesavaraja ²

Affiliations
1 Department of Master of Computer Applications, Dr. Sivanthi Aditanar College of Engineering, Tamil Nadu, IN
2 Department of Computer Science and Engineering, Dr. Sivanthi Aditanar College of Engineering, Tamil Nadu, IN

Abstract

Grid Computing provide ample opportunities in many areas. A authorized user should be able to access any data that has been created by him, in the Grid, considering all the factor for security and other vital criteria such as reliability, availability, secure reading and writing of data. The proposed service helps the authorized user to create, modify, view the data that has been created by him or his group, irrespective of whether he is inside the grid environment or he is outside the grid. This service is maintained and moderated by a coordinator which takes care of factors of security, other vital criteria and properties of distributed and grid environment. In order to increase the availability of data LTSIA uses a replication algorithm that decides the number of replicas that has to be created and where to store the replicated data. The decisions are made from the information stored in the service table linked with co-ordinator. Elliptic curve crypto Algorithm is used to encrypt data while being transmitted outside the grid. The authorized user can use a unique id assigned to him to decrypt the data or shall use a digital certificate system as a tool to decrypt it.

Keywords

Availability, Cloud Computing, Grid Computing, Location Transparency, Replication, Security.

Full Text

     

D. Kesavaraja ¹, R. Balasubramanian ², D. Jeyabharathi ³, D. Sasireka ⁴

Affiliations
1 Department of Computer Science and Engineering, Dr Sivanthi Aditanar College of Engineering, Tiruchendur, Tamilnadu, IN
2 Department of Computer Science and Engineering, Manonmaniam Sundarnar University, Abishekapatti, Tirunelveli, Tamilnadu, IN
3 Department of Computer Science and Engineering, Einstein College of Engineering, Tirunelveli, Tamilnadu, IN
4 Department of Information Technology, PSN College of Engineering and Technology, Tirunelveli, Tamilnadu, IN

Abstract

Cloud computing provides ample opportunity in many areas such as fastest image transmission, secure and efficient imaging as a service. In general users needs faster and secure service. Usually Image Compression Algorithms are not working faster. In spite of several ongoing researches, Conventional Compression and its Algorithms might not be able to run faster. So, we perform comparative study of three image compression algorithm and their variety of features and factors to choose best among them for cluster processing. After choosing a best one it can be applied for a cluster computing environment to run parallel image compression for faster processing. This paper is the real time implementation of a Distributed Image Compression in Clustering of Nodes. In cluster computing, security is also more important factor. So, we propose a Distributed Intrusion Detection System to monitors all the nodes in cluster . If an intrusion occur in node processing then take an prevention step based on RIC (Robust Intrusion Control) Method. We demonstrate the effectiveness and feasibility of our method on a set of satellite images for defense forces. The efficiency ratio of this computation process is 91.20.

Keywords

Compression, Cluster, JPEG, Distributed Computing, Security.

Full Text