Browsing by Author "Bhagtya, P."

Now showing 1 - 3 of 3

Clique Displacement: A New Layout Technique
(Institute of Electrical and Electronics Engineers Inc., 2019) Bhagtya, P.; Tilwe, V.; Chandrasekaran, K.
A graph allows the representation of data in a comprehensible way either for visualizing clusters in a large data set or analyzing trafc from network devices represented as nodes and links. It has always been a good mode for understanding a problem and subsequently analyzing the solution. A graph data structure is a good choice to represent data if it can be converted in the form of nodes and edges. There are different types of problems in various elds which require a graphical method to understand the problem, and so there are several approaches for doing it. On similar lines, this paper proposes a new visualization technique for a dense graph. It is evident that existing visualization techniques, such as force-directed placement would not give optimal results for a dense graph. The proposed work is interesting and has several potential applications such as in analyzing graphs for Social Networks, Biological applications, etc. There are many graph layout techniques available, and it has been studied for years. Every year some new method is proposed, or an older one is improved. This is because of the exponentially increasing data, which requires a better layout technique for representation. Today, graph is used in different research areas due to its simplicity in the way of representing the information. A graphical representation is always good for understanding the information easily. When these graphs become dense, the overlapping of edges in the graph also increases by a certain amount, and it becomes difcult to understand the graph or tough to visualize the graph. For a better understanding of dense graph, this paper proposes a solution by dividing the original graph into sub-graphs forming cliques. These cliques are then aligned in special positions to overcome the confusion between the connectivity of nodes. This idea considerably solves the visualization problem with dense networks, and the results show a better visual representation. Â© 2019 IEEE.
Singlow: Simulator for General Network Flow Problems
(Institute of Electrical and Electronics Engineers Inc., 2020) Raghavan, S.; Bhagtya, P.; Chandrasekaran, K.
Simulation is an important process and an inevitable part of engineering. There are several applications of simulation with one of the important being visualization of complex methods and processes. This paper aims at creating a simulator for general network flow optimization problems. This work uses a modular approach for creating a simulator. A simulator in this area is necessary for several reasons. The main reason for requirement is its usefulness in explaining the problems to the people/students who might use these kinds of network optimization methods to solve several variety of problems. This simulator can simulate standard problems namely transportation problems with various methods, network flow problem and some popular problems in that area. This simulator will be helpful for educational institutions to teach the students about the standard problems on network flow optimization. Here this paper proposes a framework i.e. Singlow for the above mentioned purpose. This paper explains the framework with the flow of execution by keeping in mind a general simulation software. The Simulator has been designed and implemented using Processing 3.4, a software which facilitates designing graphical user interfaces. Â© 2020 IEEE.
Workload classification in multi-vm cloud environment using deep neural network model
(Association for Computing Machinery, 2021) Bhagtya, P.; Raghavan, S.; Chandrasekaran, K.; Divakarla, U.
In this competitive world, everyone needs to be prepared for future risks and emergency conditions. In a multi-cloud environment users can easily shift from one cloud to another cloud because of the available data and application transfer technologies. Therefore a strong forecast system is mandatory for such conditions and to stop user migration to other clouds. Virtual Machine (VM) plays an important role in effective resource management and cost reduction in cloud infrastructure. Workload prediction in multi-VM is very useful to handle uncertain situations. In this paper, we propose a promising workload prediction technique that can handle the workload from multiple virtual machines. It has a pre-processing and feature selection engine that handles direct data from these virtual machines and the model is strong enough in classifying data based on historical workloads. This classification enables extra knowledge for the cloud vendor to optimize resource usage. This strategy can be used for producing an alarm whenever there is continuously high utilization of resources in the future. Here, our prediction methodology is experimented with a popular real-world Grid Workload Archive (GWA) dataset and it achieves more than 85% prediction accuracy for CPU, Memory and Disk Utilization. Â© 2021 Owner/Author.