Best Online Journal of Computer Science and Technology - IRA-JOURNAL

International Research & Analysis Journal facebook twitter youtube


IRAJ Volume 9 Issue 1, 2016-04-15 To 2016-06-14

  • pages (214-220)

    Reads (85)

    Topic Name: Ranking Based System for Delay Tolerant Network (DTN) Using AODV Routing Protocol

    Author (s) : Nikeeta Choubey, Sandeep Sahu

    Keywords : Delay Tolerant Network, Malicious Nodes, Ranking System, AODV Routing.

    Quick abstract

    DELAY-TOLERANT NETWORKS (DTNs) are relatively new class of networks, wherein sparseness and delay are particularly high. DTN are characterized with the Heterogeneity, intermittent connectivity between adjacent nodes, lack of end-to-end links, high delays and error rates. These characteristics pose several challenges in DTNs. Especially, malicious and selfish behavior of node represents a serious threat against routing in Delay Tolerant Networks (DTNs).Due to presence of malicious and selfish node DTNs is regarded as great challenge. In our work we proposed a Ranking Based System, detecting the behavior of the nodes and thus increase the reliability of the nodes. The basics idea of Ranking Based System is to judge the node based on the process of transmitting message from one node to another. A Ranking of every node is being proposed to be added and use along with the AODV Routing Protocol. For execution and comparison the proposed system is simulated through Network Simulator2 (NS2)

  • pages (221-226)

    Reads (84)

    Topic Name: High Performance Trust Mechanism Using AODV Protocol in Manet

    Author (s) : Alka Choubey, Sandeep Sahu

    Keywords : MANET, Trust value, malicious nodes, AODV, routing, congestion, delays.

    Quick abstract

    MANET is an infrastructure-less network in which various nodes comes nearby to each other and forms a network. The major problems in MANET are the route discovery which causes slower communication, and security because of malicious node which acts negatively. One of the most popular routing protocols that is vulnerable to different attacks during the route discovery. It is ad hoc on demand distance vector (AODV) routing protocol. Our thesis work focuses in solving the problem of malicious activities & also slower communication occurring due to possible solutions. We apply trust mechanism which means believe of a node to keep reliability factor. In previous work, for calculating value of trust, nodes perform lot of calculations and transfer of packets among them. This thesis work proposes a mechanism to calculate the trust value during the course of data transfer over the network, so as to avoid any possible delays and congestion over the network. We have considered packet delivery ratio, end to end delay and jitter value. We used Network Simulator-2.32 for evaluation of these parameters.

  • pages (227-233)

    Reads (86)

    Topic Name: Resource Allocation in Cloud Environment: Challenges and Opportunities

    Author (s) : Pragati Patel, Alekh Dwivedi, Dr. Vineet Richariya

    Keywords : Cloud Computing, Virtualization, VM, Dynamic resource Provisioning, Adaptive Threshold;

    Quick abstract

    In the present scenario cloud computing becomes a popular concept; it can be viewed as a repository hosting services and data in a very large scale. Cloud computing offers information Storage, information processing and delivery of services with the help of internet anywhere, from any location in the world. On demand service delivery makes cloud computing more popular, causing the researchers to focus on it. Management of application running in cloud architecture created a major challenge of On-demand resource provisioning and adaptability to the dynamically changing workloads. Existing methods are emphasizing on migration of VM and allocation of physical resource to VM Resulting suspension of cloud application due to mandatory shutdown of associated VM migration. For the purpose of better resource utilization in cloud environment we can apply resource allocation at the application level, in-spite of mapping physical resource to VM at physical level. Threshold based resource allocation schemes can be used to enhance performance of cloud resources, which dynamically allocates resources according to load changes (to fulfil peak load demand). The experimental results in CloudSim toolkit shows threshold based resource allocation scheme can improve resource utilization and can reduce user usage cost.

  • pages (234-241)

    Reads (75)

    Topic Name: Efficient Source Routing using Table based Routing Updates

    Author (s) : Shivani Goyal, Ashok Verma

    Keywords : Wireless Networks, Secured Routing, DSDV, Packet Delivery Ratio, Security, Key Distribution

    Quick abstract

    RESEARCH on security of MANETs remains active, in spite of years of exploration, in both academia and industry. It is partially due to the fact that no mature solution is widely accepted and the growing availability of small, personalized mobile devices with peer to peer communication capability through wireless channels. The characteristics of Mobile Ad-hoc NETwork (MANET) such as node mobility, dynamic infrastructure, unreliable multi-hop communication channel, resource limitation and physical vulnerability, securing MANET has made more challenging. This work proposes to further enhance the work done by Elhadi et al [1], who have suggested using MRA (Malicious Report Authentication) and digital signature to provide the details of false alerts generated by malicious nodes and key authentication before communication. This work uses CTS and RTS signals to exchange the key and symmetric cipher algorithm to encrypt and decrypt the data. Proposed implementation shall be tested with DSDV routing protocol which can be further enhanced and tested on other routing protocols.

  • pages (242-248)

    Reads (79)

    Topic Name: Reliable Small Application Migration for Hadoop

    Author (s) : Leelavati Patel, Deepak Choubey

    Keywords : Hadoop, Cloud Computing, Distributed Computing, MapReduce, Small Files Management

    Quick abstract

    Hadoop has two core components: HDFS to store large data with reliability and MapReduce is a programming model which processes the data in parallel and distributed manner. Hadoop does not perform well for small files as a large number of small files pose a heavy burden on the NameNode of HDFS and an increase in execution time for MapReduce is encountered. [1] Hadoop is designed to handle huge size files and hence suffers a performance penalty while dealing with large number of small files. This research work gives a new approach to handle small files over Hadoop with less performance penalties. In proposed approach, merging of small file is done using MapReduce programming model on Hadoop. This approach applies the clustered approach for merging of the files so that merged files shall not be of unexpectedly greater size. The clustered approach will also use less amount of memory on MapReduce Server after margining of files hence will improve the performance of Hadoop.