Kannan Govindarajan

CLOUDRB: A framework for scheduling and managing High-Performance Computing (HPC) applications in science cloud

In recent years, the Cloud environment has played a major role in running High-Performance Computing (HPC) applications, which are computationally intensive and data intensive in nature. The High-Performance Computing Cloud (HPCC) or Science Cloud (SC) provides the resources to these types of applications in an on demand and scalable manner. Scheduling of jobs or applications in a Cloud environment is NP-Complete and complex in nature due to the dynamicity of resources and on demand user application requirements. The main motivation behind this research study is to design and develop a CLOUD Resource Broker (CLOUDRB) for efficiently managing cloud resources and completing jobs for scientific applications within a user-specified deadline. It is implemented and integrated with a Deadline-based Job Scheduling and Particle Swarm Optimization (PSO)-based Resource Allocation mechanism. Our proposed approach intends to achieve the objectives of minimizing both execution time and cost based on the defined fitness function. It is simulated by modeling the HPC jobs and Cloud resources using the Matlab programming environment. The simulation results prove the effectiveness of the proposed research work by minimizing the completion time, cost and job rejection ratio and maximizing the number of jobs completing their applications within a deadline and meeting the users satisfaction. The proposed work has been tested in our Eucalyptus-based cloud environments by submitting real-world HPC applications and observed the improvements in performance.

DDoS defense system for web services in a cloud environment

Abstract Recently, a new kind of vulnerability has surfaced: application layer Denial-of-Service (DoS) attacks targeting web services. These attacks aim at consuming resources by sending Simple Object Access Protocol (SOAP) requests that contain malicious XML content. These requests cannot be detected on the network or transportation (TCP/IP) layer, as they appear as legitimate packets. Until now, there is no web service security specification that addresses this problem. Moreover, the current WS-Security standard induces crucial additional vulnerabilities threatening the availability of certain web service implementations. First, this paper introduces an attack-generating tool to test and confirm previously reported vulnerabilities. The results indicate that the attacks have a devastating impact on the web service availability, even whilst utilizing an absolute minimum of attack resources. Since these highly effective attacks can be mounted with relative ease, it is clear that defending against them is essential, looking at the growth of cloud and web services. Second, this paper proposes an intelligent, fast and adaptive system for detecting against XML and HTTP application layer attacks. The intelligent system works by extracting several features and using them to construct a model for typical requests. Finally, outlier detection can be used to detect malicious requests. Furthermore, the intelligent defense system is capable of detecting spoofing and regular flooding attacks. The system is designed to be inserted in a cloud environment where it can transparently protect the cloud broker and even cloud providers. For testing its effectiveness, the defense system was deployed to protect web services running on WSO2 with Axis2: the defacto standard for open source web service deployment. The proposed defense system demonstrates its capability to effectively filter out the malicious requests, whilst generating a minimal amount of overhead for the total response time.

Particle Swarm Optimization (PSO)-Based Clustering for Improving the Quality of Learning using Cloud Computing

Virtual Learning is a key enabler for giving equal opportunity to all throughout the globe. However, the pedagogical approach preferred by a group of learners may differ from another set of learners. By providing different pedagogical approaches through virtual learning, it is possible to satisfy the need of the learners, thereby improving the quality of learning. To identify the preference or choice of the pedagogy, the behavior of the learners is captured and analyzed. According to the understanding capability, the appropriate pedagogy is adopted for that learner. The conventional Learning Management System (LMS) plays a major role for achieving effective teaching and learning process. However, the conventional LMS fails to address the effective teaching and learning process by not providing the contents based on individual users ability. The proposed work mainly intends to capture the data from students, analyze and cluster the data based on their individual performances in terms of accuracy, efficiency and quality. The clustering process is carried out by employing the population-based metaheuristic algorithm of Particle Swarm Optimization (PSO). The simulation process is carried out by generating the data. The generated data is based on the real data collected from engineering undergraduate students. The proposed PSO-based clustering is compared with existing K-means algorithm for analyze the performance of inter cluster and intra cluster distances. Finally, the processed data is effectively stored in the Cloud resources using Hadoop Distributed File System (HDFS).

An architectural framework to solve the interoperability issue between private clouds using semantic technology

Cloud Computing is the blending technology of parallel and distributing computing paradigm, and its main aim is providing Everything as a Service (XaaS) to the consumers. The Infrastructure as a Service (IaaS) is one type of service models in Cloud Computing to provide infrastructure to the customers. The private cloud is one of the deployment models in cloud computing, and it offers resources to the consumers in an on-demand manner within or inside the organization. It is managed by different type of cloud middlewares and each them have their own mechanism and protocols. So there is an interoperability issue between these different type of private clouds, to solve this issue it is essential to describe a common mechanism. In this paper, we have devised a a broker based architectural framework to solve the interoperability issue between the Eucalyptus and OpenNebula based private clouds. The proposed work has incorporated with semantic based resource description and discovery, capacity based resource selection and resource provisioning mechanism. The proposed framework solves the interoperability between private clouds, enhances the efficiency of cloud resources, increases the scheduling success rate resulting in increase of throughput of application requests submitted to the resource broker in a better manner.

Intelligent semantic discovery in virtualized grid environment

A basic function of Grid Scheduler is to decide which jobs to run on what computational resources based on the users application requirements. Since dynamic configuration of the execution environment is not supported by the grid middleware. By incorporating the virtualization technology in the existing grid infrastructure, the grid has been made configurable by creating the virtual resources based on the application requirements of the user. To create the virtual resource, it is essential to identify the suitable physical resource in which the virtual resource is created. In this paper we devise an intelligent semantic discovery mechanism to identify the suitable physical resources for the creation of virtual resources. In our proposed work physical as well as virtual information has been semantically represented using ontology knowledge base. The proposed intelligent discovery mechanism helps the Care Resource Broker [1] for making intelligent scheduling decisions in the submission of jobs to physical as well virtual resources based on the application requirements followed by the job submission. The improved performance by means of higher throughput is the result of the proposed work.

Dynamic Learning Path Prediction — A Learning Analytics Solution

In the course of the last few years, many educational and research communities have been deeply invested in the development of learning analytics. Learning analytics measures the effectiveness and efficiency of learning environments, in order to understand the needs of learners and to improve the teaching process. The research presented in this paper uses Parallel Particle Swarm Optimization (PPSO) mechanism to analyze and predict a dynamic learning path for learners based on competence and meta-competence values observed in a learning environment. The proposed system is able to auto-configure and auto-customize itself to offer personalized and individualized instruction, and calculate an optimal learning pathway for learners. Furthermore, it provides on-demand and adaptive support for learners based on their needs. Experimental evaluations – carried out within a Java Programming course -- demonstrate the effectiveness of the proposed system.

Failure-aware resource provisioning mechanism in cloud infrastructure

Cloud resource providers provide the compute, storage and network resources in an on demand manner from cloud infrastructure. Resource Provisioning is one of the biggest challenging tasks in the cloud infrastructure, due to the dynamic nature of cloud resources and on demand user application requirements. However, in the cloud infrastructure, failure is one of the most essential factors that plays a main role in the completion of jobs within the user-specified deadline and minimize the execution cost. The proposed failure-aware resource provisioning mechanism involves three stages: (1) Predict the availability of cloud resources (2) Cluster the cloud resources based on the availability (3) Optimal Selection of Clustered resources. The proposed approach is integrated with the Cloud Broker to allocate the cloud resources for provisioning in an optimal manner. It is evaluated by performing the various simulation experiments. The simulation results prove the effectiveness in terms of makespan, execution cost and failure rate of the jobs.

CDM Server: A Data Management Framework for Data Intensive Application in Internal Private Cloud Infrastructure

The idea of cloud computing has been evolving over the years, towards the direction that best suits the perpetual needs of organizations. The majority of organizations have started to house their own Internal private cloud, rather than relying on third party private clouds for wide range of reasons. Development of an internal private cloud has to overcome a wide range of challenges to match the performance of public cloud solutions. The major challenge is the development of a framework for efficient data management for processing the Big Data for data intensive applications in this private cloud infrastructure. We have proposed a data management framework called CDM server which handles these data management challenges effectively. The CDM server provides a platform with the key requirements of cloud, namely increased scalability, Big Data handling, maximizing cluster utilization and performing capital scheduling of data intensive jobs.

Virtual machine placement optimization in SDN-aware federated clouds

The cloud providers provide computing, storage and network resources in an on-demand manner and pay as per usage mode. Nevertheless, the cloud consumers are facing issue to find out the suitable cloud resources that will meet their application requirements. Hence, the proposed research work introduces the cloud brokering concept that acts as the mediator between the user and the cloud provider to perform the mapping between the user application requirements with the available cloud resources. In addition to that, in this research paper, we introduced Software-Defined Networking (SDN) based networking to manage and configure the networks in a dynamic manner. However, the conventional cloud brokering concepts selects the resources from a single cloud provider; in some circumstances, it fails to satisfy the user requests in a single cloud provider. Henceforth, the cloud broker selects the cloud resources across the multiple cloud resources known as federated clouds. In these scenarios, the placement of virtual machines is the most challenging and complex issue, hence, in this research work, a Multiple Knapsack Problem (MKP) has been designed to solve the complex virtual machine placement problem in the SDN-aware federated clouds. It is simulated and tested based on the real-world application traces and the various performance metrics such as execution time, execution cost, and user satisfaction values are measured.

VCDE: a toolkit for virtual cluster creation for grid environment

Virtualisation is a promising technology that has attracted much attention, particularly in the grid community. Recent advances using virtualisation technologies enable multiplexing the physical resources by means of virtual machines (VMs) resulting in better resource utilisation. We propose virtual cluster development environment (VCDE) to dynamically form virtual cluster on demand, providing the grid execution environment using VMs. The formation of a grid environment to execute any parallel or sequential application has been automated by VCDE. The user can simply submit the job along with the requirements of hardware and software needed for the execution. The VCDE provides easy access to the grid by reducing the burden of Globus toolkit installation and other configurations. It is important to note that there is no human intervention in the preparation of virtual cluster as well as the grid environment. The proposed VCDE uses virtualisation technologies, and it helps the user to make use of grid as easily as possible.