S. Vigneshwari ¹, M. Aramudhan ²

Affiliations
1 Department of Computer Science and Engineering, Sathyabama University, Chennai, IN
2 Perunthalaivar Kamarajar Institute of Engineering and Technology, Karaikal, Tamilnadu, IN

Abstract

Ontology is the best way for representing the useful information. In this paper, we have planned to develop a model which utilizes multiple ontologies. From those ontologies, based on the mutual information among the concepts the taxonomy is constructed, then the relationship among the concepts is calculated. Thereby the useful information is extracted. There is multiple numbers of ontologies available through the web. But there are various issues to be faced while sharing and reusing the existing ontologies. To resolve the ambiguity which exists, when comparing two concepts are semantically similar, but physically different, an approach is proposed here to index and retrieve the documents from two different ontologies. The ontologies used are WordNet and SWETO ontology. The results are compared based on semantic annotation based on RMS and hashing between the cross ontologies using Rabin Karp fingerprinting algorithm. Also the datasets are trained to yield better results.

Keywords

Concept Similarity, Information Extraction, Hashing, Ontological Relationship, Semantic Annotation, Training the Ontology.

Full Text

   

R. K. Nadesh ¹, Meduri Jagadeesh ¹, M. Aramudhan ²

Affiliations
1 SITE, VIT University, Vellore - 632014, Tamil Nadu, IN
2 Department of IT, PKIET, Puducherry - 609603, IN

Abstract

The demand for cloud based resources has been increasing, catering to the need for enormous growth in Cloud data centers to serve its customers by providing different types of VMs and providing support too many business applications is always challenging. Cloud data center is the collection of physical resources, which provide services to its customers in a virtual manner. There is need for effective scheduling algorithms along with scheduling maintenance activity of the servers. To provide continuous and uninterrupted services to its customers, data centers is to be maintained properly by looking at the issues like scheduling the server’s resources to requests, queuing the requests to the servers and proper maintenance time allocated to each server. Devising the algorithm and implementing is done in algorithmic perspective with Open Nebula toolkit and the results are evaluated using the sunstone GUI server. To enable resources in Open Nebula toolkit, ruby scripts are coded with embedded C. Open Nebula CLI commands are used to check the performance of the proposed design.

Keywords

Command Line Interface (CLI), Virtual Machine (VM)

Full Text

   

R. Charanya ¹, M. Aramudhan ², Ra. K. Saravananaguru ¹

Affiliations
1 VIT University, Vellore - 632014, Tamil Nadu, IN
2 PKIET, Karaikal - 609603, Pondicherry, IN

Abstract

Personal Health information is highly confidential. The Main objective is to find out the techniques to protect the health care data from unauthorized user in cloud. Due to privacy concern the personal data are need to be protected from third party and also from unauthorized user. In existing work they used cryptographic techniques like KP-ABE, Attribute based encryption, CP-ABE, etc. This paper addresses the challenges in each access control techniques, and its pros and cons. Future work is to overcome the challenges in existing techniques by using Keyless signature algorithm by using hash function.

Keywords

Access Control Models, Encryption Techniques.

Full Text

   

S. Mourougan ¹, M. Aramudhan ²

Affiliations
1 Periyar University, Salem - 636011, Tamil Nadu, IN
2 Department of IT, Perunthalaivar Kamarajar Institute of Engineering and Technology, Nedungadu, Karaikal - 609603, Puducherry, IN

Abstract

Background/Objectives: In current days, administrating security effectively for computer resources has become a difficult task for the administrator. One of the security problems is Denial of Service (DoS), is a type of attack that tries to prevent legitimate users from accessing either the services or resources, by generating large number of artificial packets send towards the victim resource. In turn, the victim resource is unable to extend the service to legitimate user. To meet this type of attacks, numerous detecting and preventing systems have been proposed, but they were suffering from low detection accuracy and high false alarms. Methods/Statistical Analysis: A new computational technique was proposed to perform the classification task andextracting features from KDDCUP 99 datasets using genetic algorithm and Particle Swarm Optimization. This research work focuses on the identification of DoS attacks with high detection accuracy and less false alarms. Findings/Conclusion: In the proposed Intrusion Detection System model, attacks are identified by training the Particle Swarm Optimization classifiers with Genetic-Particle Swarm Optimization based on wrapper feature selection which is superior to those classical intrusion feature selection. The proposed work was implemented in MATLAB. The result shows high detection accuracy and fewer false alarms compared to the existing models.

Keywords

Denial of Service, Genetic Algorithm, Intrusion Detection System, Particle Swarm Optimization

Full Text

   

S. Jayapraksh ¹, M. Aramudhan ²

Affiliations
1 Bharathiyar University, Coimbatore - 641 046, Tamil Nadu, IN
2 Department of Information and Technology, PKIET, Karaikal - 609 603, Puducherry, IN

Abstract

Background/Objective: An objective of reliable ranking method to encourage the growth of Cloud Computing. The challenges of probability ranking are various natures of cloud services and complexity of malicious ratings. Methods/Statistical Analysis: The proposed Classical Probability Ranking Principle (CPRP) is used to select the matched providers, rank it and choose the optimal provider for the service in federated cloud. In all situations, federated cloud presents list of provider to the users, with reference to which system to choose and allocate the best to the user. Each provider is associated with SMI attributes, based on attributes, an optimum ranking of the choices can be derived using CPRP. Findings: This chapter is discussed the new ranking mechanism for ranking the providers using classical probability ranking on the basis of the SMI attributes. All the shortlisted providers are examined and then ranking is done on the basis of the present and past values of SMI. Applications/Improvements: The future work on ranking manages with variation in QoS attributes that is presentation by adopt the evolutionary algorithms. This work also extends the feature model to non-theoretical QoS attributes.

Keywords

Matched providers, optimal provider, Ranking, SMI attributes, SLA, CPRP.

Full Text

   

R. Rajeshkannan ¹, M. Aramudhan ²

Affiliations
1 Faculty of Computing Science and Engineering, VIT University, Vellore - 632014, Tamil Nadu, IN
2 PKIET, Karaikal - 609603, Puducherry, IN

Abstract

Background: Cloud computing is an emerging technology in a business. It is used to access the application or services and infrastructures at anywhere any time. Load balance means that to share the work across the multiple computing resources to serve higher for the user and utilize the resource with efficiency for reach the good performance of the application. These ideas are enforced with software system, hardware or each. Statistical Analysis: The various load balancing algorithms are compared with quality of service parameters in a cloud network. This analysis helps to identify the effective load balancing algorithm for optimizes resource use, maximizes throughput, minimizes response time, and avoids overload. Findings: The load balancer is indorsed in all situations for to supply service continuity and handling additional traffic. Therefore the effective load balancing algorithms needed to form economical resource utilization by provisioning of resources to cloud user's on-demand basis. Application: This paper discusses numerous load balancing algorithms so as to improve resource utilization and quality of services in cloud computing environment.

Keywords

Cloud Computing Model, Cloud Computing Characteristics, Load Balancing, Task Scheduling, Virtual Machine.

Full Text

   

S. Mourougan ¹, M. Aramudhan ²

Affiliations
1 Periyar University, Periyar Palkalai Nagar, Salem - 636011, Tamil Nadu, IN
2 Department of IT, Perunthalaivar Kamarajar Institute of Engineering and Technology Nedungadu, Karaikal - 609 603, Puducherry, IN

Abstract

Background/Objectives: Federated cloud architecture is heterogeneous and distributed model that provides infrastructures related to cloud by aggregating different IaaS providers. In this position, it is an exciting task to pick the optimal cloud provider for the services and deploy it in affordable cost. In this paper, an effective new methodology for establishing trust and the different cloud providers based on the trust value using SMI attributes based on regression tree model is to enrich security and privacy in the federated cloud. Methods/Statistical Analysis: The scoring is computed based on the depth of the traversals in regression tree. Policy based trust, SLA verification trust; reputation trust and Evidence based trust are proposed in the architecture to build secured federated cloud. Findings/Conclusion: Broker trust level can also be identified and classified into four different categories such as completely trusted, partially trusted, minimally trusted and not trusted. Simulation results show that the performance of the ranked based mechanism depends on its implementation and better than without rank based federated cloud. Applications/Improvements: Will focus on mathematically formal frameworks for reasoning about trust, including modeling, languages and algorithms for computing trust.

Keywords

Federated Architecture, Policy Trust, Regression Tree, SLA, SMI.

Full Text

   

L. Aruna ¹, M. Aramudhan ²

Affiliations
1 Periyar University, Salem - 636011, Tamil Nadu,, IN
2 Department of IT, PKIET, Karaikal - 609603, Puducherry, IN

Abstract

Objectives: To improve Quality of Services (QoS) and ranking cloud service providers in federated cloud environment. It is also aimed to resolve various issues faced by the user and providers in cloud. Methods: In order to avoid the workload of existing cloud model and to compete with the well-known cloud service providers, customized federated cloud provider architecture was suggested. It is suggested to manage and to maintain QoS for the submitted tasks. The proposed architecture consists of external world, middle world and internal world. Each world plays vital role for customizing the request in federated environment to enhance QoS. To avoid starvation in the architecture to suggest the differentiated module and resolves the various key issues using the Stochastic Markov process model and also evaluating the cloud providers based on the quality of service requirements. Findings: Service Level Agreement (SLA) in single and federated cloud plays a vital role in enhancing Quality of Services. Single cloud service provider representation does not provide the skilled QoS, when workload becomes high. Customized federated cloud architecture reduces the drawbacks of single cloud service provider. The proposed architecture was implemented in CloudSim using Java. The simulation result of the proposed architecture proves enhanced QoS to the user and cloud service providers. The proposed federated cloud model enhances the Quality of Services by more than 18% of existing single and federated cloud model. The parameters considered for the simulation for numbers of users, providers, load factor, turnaround time and average load deviation of tasks. Applications: This architecture can be used to rank different cloud service providers and it can be trusted for any distributed cloud services with extended QoS.

Keywords

Cloud Sim, QoS and Federated Architecture, Stochastic Markov, SLA.

Full Text

   

Anusuya Kirubakaran ¹, M. Aramudhan ²

Affiliations
1 Mother Teresa Women’s University, Kodaikanal - 624101, Tamil Nadu, IN
2 PKIET, Karaikal - 609603, Puducherry, IN

Abstract

Background: In the competitive data driven business world, business Intelligence (BI) team converts the raw operational data to information for decision making. Operational system captures the day-to-day operations and BI database refreshes operational data periodically. Methods: A component to create the metadata repository which maintains the current BI database summary by logical data partitioning using range based partition for frequently changing parameter which are critical to business. During different time frequency, using metadata repository component identifies the latest data victim between BI vs operational data and refreshes the modified victim to BI database. Findings: In traditional data loading approach from Operational system to BI database, huge volume of data gets refreshed periodically irrespective of modifications, which leads to higher processing time and cost. To overcome this limitation, this proposed methodology helps to identify the latest data victims present in operational systems instead of bulk data replacement which can minimize the processing time and enables faster data transformation to achieve "Time to Decision" and "Quick to Market" implementation for business enhancements. Also component can be scheduled for data refresh with different time frequency for multiple critical to business as well as frequently changing parameters. Applications: In financial, traffic, weather, e-business, Logistics&stock management transactions, data changes frequently and process big data periodically to gain real-time knowledge discovery for time sensitive decision making.

Keywords

Frequently Changing Data,Time-sensitive Business Intelligence.

Full Text