“Trust based Proactive Resource Provisioning in Clouds” Sanctioned by U.G.C. as major project costing Rs. 10,48,000/- with One Research Fellow (Completed).
Abstract: Cloud computing has become an innovative computing paradigm, which aims at providing reliable, customized, Quality of Service (QoS) and guaranteed computing infrastructures for users. But the initialization of a new virtual instance causes a several minutes delay in the hardware resource allocation. Furthermore, cloud provides a fault tolerant service to its clients using the virtualization. But, in order to attain higher resource utilization over this technology, a technique or a strategy is needed using which virtual machines can be deployed over physical machines by predicting its need in advance so that the delay can be avoided. To address these issues, a value based prediction model is proposed in this report for resource provisioning in which a resource manager is used for dynamically allocating or releasing a virtual machine depending upon the resource usage rate. In order to know the recent resource usage rate, the resource manager uses sliding window to predict the system behavior in advance. By predicting the resource requirements in advance, a lot of processing time can be saved. Earlier, a server has to perform all the calculations regarding the resource usage that in turn wastes a lot of processing power thus decreasing its overall capacity to handle the incoming request. The main feature of the proposed model is that a lot of load is being shifted from the individual server to the resource manager as it performs all the calculations and therefore the server is free to handle the incoming requests to its full capacity.
One of the main aim of cloud computing is to provide bigger data center that will cater the storage needs of end user. In a data centre, server clusters are used to provide the required processing capability to get acceptable response time for interactive applications. In this work, an interactive system based on queuing model is presented in which the cloud customer (CC) initially establishes the session to access the resources. The proposed model uses banker’s algorithm for resource allocation due to which deadlock for resource allocation among various processes is not possible. Moreover, by putting restriction on number of login users, resources are not choked out even in case of heavy demand of resources. The concept of resource allocation matrix helps the cloud service provider to predict the resource requirements in advance. Since new sessions can be established and existing sessions can be terminated, the number of logon users can change over time. Resources are dynamically allocated according to the requirements of the user. The results obtained are accurate in terms of predicting the minimum number of processor nodes required to meet the performance goal of an interaction application.
Resource provisioning is a critical issue in the environment of heterogeneous clouds because storing sensitive data on public cloud is not secure whereas managing a private cloud is not so cost effective. Using only private cloud makes system prone to deadlock which can not easily detected using traditional techniques. In this report, a deadlock manager is proposed which uses Social Network Analysis (SNA) techniques for deadlock detection in resource allocation graph. Four types of clouds are used collectively to remove if any deadlock is detected. A new concept of free space cloud is proposed in this research report which helps in utilization of resources more efficiently on private cloud. A set of rules are proposed for a Resource Pool Manager (RPM) to increases the utilization of resources in private cloud and decrease the response time of request in case any deadlock is detected. Selection among clouds is done by assigning priorities to the requests and providing resources accordingly from different clouds. The performance of Resource Pool Manager is evaluated by using Cloudsim and resource utilization comes out to be very good in our proposed technique.
“Big Data, Internet of Things and Cloud Computing: Architecture, Issues and Applications in Indian Perspective”, sanctioned by Department of Electronics and Information Technology (DeitY) costing 14,76,000/- with One Research Fellow (Ongoing).
Works completed w.r.t. the proposed works
Big data streams are generated continuously at unprecedented speed by thousands of data sources. The analysis of such streams need cloud resources. Due to growth of big data over cloud, allocating appropriate cloud resources has emerged as a major research problem. The current methodologies allocate cloud resources based upon data characteristics. But due to random nature of data generation, the characteristics of data in big data streams are unknown. This poses difficulty in selecting and allocating appropriate resources to big data stream. Solving this problem, an efficient resource management system is proposed in this paper. The proposed system initially estimates the data characteristics of big data stream in terms of volume, velocity, variety and variability. The estimated values are expressed in terms of a vector called Characteristics of Data (CoD). On the other hand, clusters of cloud resources are created dynamically with the help of Self-Organizing Maps (SOM). SOM uses CoD to create and allocate cluster to big data stream. Moreover, the topological ordering of clusters formed by SOM is used to reduce waiting time. The proposed system is tested experimentally. The experimental results show that the proposed system not only efficiently predicts data characteristics but also effectively enhanced the performance of cloud resources.
Zika virus is a mosquito-borne disease that spreads very quickly in different parts of the world. In this article, we proposed a system to prevent and control the spread of Zika virus disease using integration of Fog computing, cloud computing, mobile phones and the Internet of things (IoT)-based sensor devices. Fog computing is used as an intermediary layer between the cloud and end users to reduce the latency time and extra communication cost that is usually found high in cloudbased systems. A fuzzy k-nearest neighbour is used to diagnose the possibly infected users, and Google map web service is used to provide the geographic positioning system (GPS)- based risk assessment to prevent the outbreak. It is used to represent each Zika virus (ZikaV)-infected user, mosquito-dense sites and breeding sites on the Google map that help the government healthcare authorities to control such risk-prone areas effectively and efficiently. The proposed system is deployed on Amazon EC2 cloud to evaluate its performance and accuracy using data set for 2 million users. Our system provides high accuracy of 94.5% for initial diagnosis of different users according to their symptoms and appropriate GPS-based risk assessment.
“Smart Disaster Management using Internet of Things (IoT) in Indian Perspective”, sanctioned by Council of Scientific and Industrial Research costing 19,62,000/- with One Research Fellow (Ongoing).