M Tech Dissertations
Permanent URI for this collectionhttp://ir.daiict.ac.in/handle/123456789/3
Browse
2 results
Search Results
Item Open Access Issues of computational resource allocation and load balancing in cloud computing using virtualization(Dhirubhai Ambani Institute of Information and Communication Technology, 2009) Bhadani, Abhay Kumar; Chaudhary, SanjayWith the advent of world wide web, the life of every person has changed drastically. No one can imagine a computer without being connected to the Internet. This dependency is likely to grow in coming years with the adoption of technologies like virtualization and cloud computing. Cloud Computing and Virtualizations one of the foremost technology which has attracted many researchers recently, which is directly going to benefit the end-users and data center service providers. It has many underlying benefits, one of them is directly related to the costs of deploying new servers. Few others are related to harnessing the power of the existing infrastructure and resources. When we look things from server side resource usage, things become quite challenging. Just to meet peak loads, high capacity servers are deployed, which remains underutilized most of the times on an average. With the help of virtualization technique, we can run multiple instances of different operating systems simultaneously. There are many virtualization tools available, which can be used to achieve maximum resource utilization. Making efficient use of computing resource (especially computational time) has always been a critical and challenging task. Several scheduling algorithms have been in place, proposed and implemented on one or other Operating Systems from time to time. But, since we have limited processors and things work in concurrent fashion, overload situation can occur hampering the overall objective, performance and throughput of the system. Things become even more challenging when it comes to distributed systems and load balancing. One of the key commercial player in the virtualization is VMWare, whereas Xen is an open-source free software. There are also few tools like Virtual Box, Linux VServer, OpenVZ, Virtuozzo and few others. Among these the most popular virtual machine monitor is Xen. The likely choice of experimenting with Xen is due to Open Source and wide acceptance by Linux community, and their continuous effort to improve, also it meets the need of industry standards at large. Consider a hypothetical scenario, where multiple instances of operating systems are running and is being used by the clients over the web in the form of cloud service. The user will use the system as if the whole server and/or system is dedicated to him. The user is completely unaware about the actual physical location of the server on which he is running his applications and disk storage space. Since, multiple OS are running on a single physical server, and multiple servers are running in the data centre. All connected via high speed network links. At some instance of time, one server may become overloaded, while other server may remain underutilized. This again poses challenge to distribute the load and make things work perfect in this situation. This situation can be handled using load balancing mechanism over the virtual machines. This thesis work tries to find a mechanism to balance the load based on computational time parameter of the virtual machines.Item Open Access Non-uniform information dissemination for performance discovery in computational grids(Dhirubhai Ambani Institute of Information and Communication Technology, 2005) Patel, Dhawal B.; Chaudhary, SanjayThe required service in any resource-sharing environments like Grid, Peer-to-Peer etc., is discovery of resources. A resource discovery mechanism returns locations of resources that match the description, given a description of resource desired. Two resource-sharing environments are well defined with respect to target communities, resources, applications, scalability and fault tolerance: Grid and peer-to-peer systems. Grids are sharing environments that rely on persistent, standards-based service infrastructure that allow location independent access to resources and services, which are provided by geographically distributed machines and networks. The design of the resource discovery approach must follow the rules imposed by the characteristics of grid environment. These characteristics are 1. Independence from central global control, 2. Support for attribute-based search, 3. Scalability, 4. Support for intermittent resource participation. Depending upon the types of resources that are shared, the grids can also be of different types, e.g. computational grids for the environment in which only computational resources are shared, data grids for the one in which data are shared. The focus of thesis is on performance discovery in computational grids. Grid schedulers, that manages the resources, requires up-to-date information about widely distributed resources in the Grid. This is a challenging problem given the scale of grid, and the continuous change in the state of resources. Several non-uniform information dissemination protocols have been proposed by researchers to efficiently propagate information to distributed repositories, without requiring flooding or centralized approaches. Recently, a new concept called the “Grid potential” proposed in, as the first step towards the design of non-uniform information dissemination protocols. In this thesis, four non-uniform dissemination protocols are analyzed for computational grids based on the concept of “Grid potential”, which follows above-mentioned requirements for resource discovery. These protocols disseminate resource information with a resolution inversely proportional to the distance of resources from the application launch point. The performance evaluation is done with respect to the dissemination efficiency and message complexity. The results indicate that these protocols improve the performance of information dissemination compared to uniform dissemination to all repositories.