Madhuri Bhavsar

Load Balancing in Grid Environment using Machine Learning - Innovative Approach

ABSTRACT Grid computing creates the illusion of a simple but large and powerful self-managing virtual computer out of a large collection of connected heterogeneous systems sharing various combinations of resources which leads to the problem of load balance. The main goal of load balancing is to provide a distributed, low cost, scheme that balances the load across all the processors. To improve the global throughput of Grid resources, effective and efficient load balancing algorithms are fundamentally important. Focus of this paper is on analyzing Load Balancing requirements in a Grid environment and proposing an algorithm with machine learning concepts to find more efficient algorithm. The decisions General Terms Grid Computing, Load balancing, Machine learning, Job migration. 1. INTRODUCTION 1.1 Grid Computing Grid is type of parallel and distributed system that enables the sharing, selection and aggregation of geographically distributed resources dynamically at run time depending on their availability, capability, performance, cost, and user quality-of -self-service requirement. Further we can also identify Grid Computing [1, 2, 4] as, a type of parallel computing that relies on complete computers connected to a network. Grids tend to be more loosely coupled, heterogeneous, and geographically distributed. In Grid computing, the details are abstracted, and the resources are virtualized.

Live Virtual Machine Migration Techniques in Cloud Computing: A Survey

Cloud computing is a service where storage and computing resources accessed on subscription basis. Cloud computing is powered by virtualization technology .Live migration is the process of moving a running virtual machine or an application between different physical machines without disconnecting the client , memory, network connectivity and storage of the virtual machine are transferred from the original host machine to the destination. This capability is being increasingly utilized in today’s enterprise environments to provide efficient online system maintenance, reconfiguration, load balancing and fault tolerance. This paper presents a detailed survey on Live Migration of Virtual machines (VM) in cloud computing.

Improving Energy Estimation based Clustering with Energy Threshold for Wireless Sensor Networks

In this paper an energy usage estimation technique (LCEFCM) has been proposed which employs the Fuzzy C-Means clustering for creating clusters in the Wireless Sensor Networks. LCEFCM reduces the energy consumption considerably compared to other clustering methods like simulated annealing and K-Means clustering. It applies the dynamic clustering mechanism combined with balanced clustering method. LCEFCM outperforms LEACHC, LEACHC Estimate(LCE) and LCEKMeans for various performance measuring factors like network lifetime, data received, alive nodes etc.

Performance analysis of local, network and distributed file systems running inside user’s virtual machines in cloud environment

Cloud computing, a recently developed paradigm, mainly focuses on resource allocation on demand. Operating Systems running in Virtual Machines can enhance their performances by adjusting resources as and when required. Due to this ever changing resource complexity, it becomes very difficult to model and analyze performance of some of the important components of Operating Systems, especially the File System. This paper presents a model, based on Queuing Theory, for performance analysis of Local, Network and Distributed File Systems running in Operating Systems of user’s VMs. This model takes into consideration parameters like average service time, average waiting time and VM migration time in file system’s performance. It also takes into consideration different failures in Cloud environment like Virtual Machine Failures, Hypervisor Failures and Communication Failures. Each File system operation is considered as a service request sent by specific Virtual Machine to the Hypervisor. The performance is evaluated based on the average time taken to service the entire request. A numerical depicting the performance analysis based on this concept has also been illustrated.

Weight Based Workflow Scheduling in Cloud Federation

To cater the need of a user as the requirement of infrastructure, application building environment or contemporary software private and public clouds are exiting for provisioning of such services. But cloud providers are facing the problem of how to deploy their applications over different clouds keeping in mind their different requirements in terms of QoS (Cost, resource utilization, execution time). Different clouds have different advantages such as one cloud will be more reliable and efficient whereas, private cloud will be more secure or less expensive. In order to get the benefits of both clouds, we can use the concept of cloud federation. When using cloud federation it becomes important how to schedule large workflows over federated clouds. Proposed work addresses the issue of scheduling large workflow over federated clouds. SMARTFED is used for the cloud federation and our algorithm is used to schedule different workflows according to the QoS parameters over the federation.

SLA Management in Cloud Federation

Now a days cloud computing is the major area of research Because cloud computing has own many benefits. Cloud computing also provides cost effective Resources so that it can become more and more helpful to IT trends. Distributed computing is an extensive arrangement that conveys IT as an administration. It is an Internet-based registering arrangement where shared assets are given like power disseminated on the electrical grid. Cloud suggest to a particular IT environment that is expected with the end goal of remotely provisioning versatile and measured IT resources. Whereas the Internet gives open access to many Web-based IT assets, a cloud is commonly exclusive and offers access to IT assets that is metered following SLAs implementation depends on guidelines that are redesigned in runtime so as to proactively recognize conceivable SLA Violations and handle them in a proper manner. Our proposed framework allows the creation and implementation of effective SLA for provisioning of service. SLA management are one kind of common comprehension between CSP (Cloud Service Provider) and customer.

Variants of Software Defined Network (SDN) Based Load Balancing in Cloud Computing: A Quick Review

Nowadays users of cloud are increasing rapidly hence handling of and allocation of that resources are the main challenge. Load balancing strategy refers to scatter the dynamic workload over the various node to guarantee that no single node is over-burden. There are few limitation of conventional load balancers in terms of flexibility and adaptability. To overcome this pitfalls, the usage of Software Defined Network based approach in load balancing proves to be advantages. Software Defined Networking is a developing innovation which helps to quickly strategies in familiarizing the administrations with the business segment without relying upon the seller based setup of the gadgets. SDN helps to set control decision for algorithm that apply for system which increases the performance of that algorithm, reduces response time, increase scalability, flexibility and results in reduction of the energy consumption of system. In this paper, we examined the feasibility of SDN-based load balancing and discussed variants of the SDN-based load balancing using various controllers.

Exhausting Autonomic Techniques for Meticulous Consumption of Resources at an IaaS Layer of Cloud Computing

Internet based computing has provided lots of flexibility with respect to the usages of resources, as per the current demand of the users, and granting them the said resources has its own benefits, if given in proper manner i.e. exactly what the user has asked. In this paper the autonomic computing concepts has been discussed which will be very useful for the better utilisation of the resources at an IaaS (Infrastructure as a Service) level of the cloud computing. As Cloud Computing is highly scalable and virtualisation has become an important means for the efficient utilisation of the resources. Seeking the right amount of the resources at right time should be the goal of any CSP (Cloud service provider), On the other hand the CSPs has to deal with the situation of over provisioning and under provisioning, there should be some self-managing scheme through which the resources should be made available to the requesting user in an efficient manner to satisfy the need of their requirement with an improved resource utilisation. We have discussed the usage of autonomic computing to enhance the resource utilisation in the IaaS of cloud computing through various ways.

Experimenting with Energy Efficient VM Migration in IaaS Cloud: Moving Towards Green Cloud

Increasing demand for Cloud infrastructure and services leads to the challenges for management and maintenance of large data Center. Data center is fully equipped with huge number of resources. Those resources consumes energy in spite of their partial or full utilization. As a result data center consumes lots of energy, which in turn increases the total cost of operation and carbon footprint in environment. These concern leads to “Green Computing”, i.e. to reduce total operational cost, Carbon Footprint in environment and efficient usage of the computing resources. In data center main processing element is virtual machine (VM), which is an instance of computing and storage resources, handles computational processes. Hence, it is important to reduce energy consumed by VM. As the workload distribution is varying in data Center as per the need, the number of VMs configured in the host are uneven, but host consumes maximum energy every time, irrespective of the workload. This leads to wastage of computational resources. This paper is intended to analyze such issues and specifically prove an algorithm which, significantly reduces energy consumption in data center, while ensuring SLA, when VM is in migration from one host to another in the data center.

Efficient Resource Monitoring and Prediction Techniques in an IaaS Level of Cloud Computing: Survey

In this paper, we have discussed about the various techniques through which the cloud computing monitoring and prediction can be achieved, This paper provides the survey of the techniques related to monitoring and prediction for the efficient usages of the resources available at the IaaS level of cloud. As cloud provides the services, which are elastic, scalable or highly dynamic in nature, which binds us to make the correct usages of the resources, but in real situations the (Cloud Service Provider)CSP’s has to face the situation of under provisioning and over provisioning, where the resources are not fully utilized and being wasted, though this is the survey paper, it ends up with the proposed model where both the concepts of the Monitoring and Prediction will be combined together to give a better vision of the future resource demand in IaaS layer of Cloud Computing.