Browsing by Author "Reddy, G. Ram Mohana"

Now showing 1 - 7 of 7

Bio-Inspired QOS Aware Resources Allocation and Management at the Cloud Data Center
(National Institute of Technology Karnataka, Surathkal, 2018) Domanal, Shridhar G; Reddy, G. Ram Mohana
Cloud comprises of many hardware and software resources and managing these resources will play an important role in executing a clients request. Now-a-days clients from different parts of the world are demanding for various services at a rapid rate. In this present situation efficient load balancing algorithms will play an vital role in allocating the clients requests and also ensuring the usage of the resources in an intelligent way so that underutilization of the resources will not occur in the cloud environment. Clients demand for different cloud resources w.r.t Service Level Agreement (SLA) in a seamless manner, therefore resource allocation and management plays an important role in Infrastructure as a Service (IaaS) based cloud environment. Computing systems in the cloud environment heavily rely on virtualization technology and thus makes the servers feasible for independent applications. Further, virtualization process improves the power efficiency of the data centers (consolidation of physical machines (PMs)) and thereby enabling the assignment of multiple virtual machines (VMs) to a single physical PM. These VM instances can be procured in the form of On-Demand and Spot instances. Consequently, some of the PMs in the cloud data center can be turned off (sleep state) and resulting in low power consumption and thus making cloud data center more efficient. In this research work, the main focus is towards designing and development of efficient QoS aware load balancing and resources allocation/management algorithms using Bio-Inspired techniques which ensures fault tolerant task execution in heterogeneous cloud environment. Experimental results demonstrate that our proposed Bio-Inspired Load Balancing and QoS Aware Resources Allocation/Management algorithms outperforms peer research and benchmark algorithms in terms of efficient utilization of the cloud resources, improved reliability and reduced average response time.
Bio-Inspired Quality of Service Aware Routing in Mobile Ad Hoc Networks
(National Institute of Technology Karnataka, Surathkal, 2014) M, Kiran.; Reddy, G. Ram Mohana
In recent years a lot of work has been done in an effort to incorporate Swarm Intelligence (SI) techniques in building an adaptive routing protocols for Mobile Ad Hoc Networks (MANETs). As centralized approach for routing in MANETs lack in scalability and fault-tolerance, SI techniques provide natural solutions through distributed approach to the adaptive routing for MANETs. The mobile nodes found in MANETs are capable of monitoring the network status as well as data processing. Thus the MANETs can be made Context Aware with the help of mobile nodes local monitoring capability. In this thesis work, a novel mobility aware bio inspired routing protocol for MANETs referred to as Mobility Aware Termite (MA-Termite) is proposed by inheriting the hill building nature of social inset Termite MA-Termite will find the reliable path between the source and destination node based on the stable nodes in terms of its mobility with the help of the local monitoring capability of nodes. Further, analytical model is also proposed for studying an asymptotic pheromone behavior of MA-Termite using two different parameters (decay rate and pheromone sensitivity) over both single and double links. The results depict how individual parameters are correlated and how they affect the global performance of MANETs. The best possible parameter values are determined for optimal performance for MA-Termite. Recently, several telecommunication applications of bio-inspired algorithms achieved remarkable success. In SI techniques, the captivating features of insects or mammals are correlated with the real world problems to find solutions. The natural question is whether it is possible to develop a new hybrid algorithm by combining the distinguishing features of these insects or mammals? In this regard, the salient features of mammals such as bats are combined with the proposed MA-Termite algorithm to come up with a new hybrid routing algorithm referred to as Bat-Termite for MANETs. Bat-Termite improved the backup route maintenance and also exhibited superior routing features such as quick route discovery, high robustness with efficient management of multiple routes and rapid route repair. One of the features of both MA-Termite and Bat-Termite algorithms is they always exclusively choose the highest pheromone link thus congests the highest pheromone link over a period of time. This undesirable behavior is referred to as stagnation. Further, MA-Termite lags in load balancing and fails to take the full benefit of multipath environment. One of the methods to avoid the stagnation problem is pheromone heuristic control. Thus, a novel heuristic hybrid Load Balanced Quality of Service (QoS) aware routing protocol referred to as Load BalancedBat-Termite (LB-Bat-Termite) is proposed for MANETs in order to solve the stagnation problem of both MA-Termite and hybrid Bat-Termite algorithms. The LB-Bat-Termite algorithm with its context awareness, QoS awareness and load balancing features exhibited considerable performance gain due to load balancing. LB-Bat-Termite produces additional control packets in order to maintain all possible paths to the destination node and thus mobile nodes spends most of its time in route maintenance than data transfer; hence causes performance degradation under high node density conditions. In prder to improve the scalability and to reduce the control packet overhead, a novel heuristic Load Balanced Termite based QoS aware routing protocol is referred to as Load Balanced-Termite (LB-Termite) is proposed for MANETs. LB-Termite exhibited considerable performance gain under both scalability and mobility factors. The proposed bio-inspired QoS aware routing algorithms in this thesis work could be used for applications such as university or campus settings, data sharing during lecturing or meeting or data sharing during virtual classrooms. The proposed algorithms for MANETs in this thesis namely MA-Termite, Bat-Termite, LB-Bat-Termite and LB-Termite are compared with the state-of-the-art bio-inspired (Simple Ant Routing Algorithm and Termite Algorithm) and non bio- inspired routing algorithms (Ad Hoc On demand Distance Vector Routing algorithm) for its performance evaluation and results are encouraging in terms of QoS parameters (Throughput, Total Packet Drops, End to End Delay and Control Packet Overhead).
Content-Based Video Copy Detection, Tracking and Identification of Movie Pirates
(National Institute of Technology Karnataka, Surathkal, 2014) Roopalakshmi; Reddy, G. Ram Mohana
Due to the exponential growth of multimedia technologies, numerous pirated contents are proliferating on the Internet and causing huge piracy as well as copyright issues. Therefore, this thesis investigates four different methodologies for combating piracy namely, Content-Based video Copy Detection (CBCD), duplicate video registration, geometric distortions computation and pirate position estimation in a movie theater. In the first methodology, this thesis targets CBCD problem, by introducing different copy detection techniques, which employ efficient video fingerprints for detecting duplicate video clips. Precisely, this research work attempts to solve some of the issues of the CBCD domain, by proposing novel video copy detection techniques, that employ color, motion activity, audio and multimodal signatures. In the second methodology, this thesis addresses the problem of video copy localization, by proposing robust registration schemes, which guarantee the accurate frame alignments of the pirate video with the master content. Specifically, this research study contributes robust temporal as well as spatio-temporal registration frameworks, which exploit visual-audio fingerprints for obtaining frame-to-frame alignments of the two video sequences. In the third methodology, this thesis aims at geometric distortions estimation, by presenting a framework using visual-audio features, which computes the geometric distortions present in the duplicate video. In the fourth methodology, this thesis attempts to emphasize the capability of video fingerprints towards the pirate position estimation problem, by performing a Case study for investigating the illegal capture location in a theater. Video copy detection, tracking, distortion estimation and pirate position approximation frameworks presented in this thesis, are evaluated with extensive experiments on different datasets. More specifically, the experimental results demonstrate the efficiency of the proposed methods over standard datasets such as TRECVID dataset, Open Video Project dataset, CC WEB VIDEO collection and real datasets comprising camcorded versions of popular movies. Further, In-Theater experiments and evaluations demonstrate the satisfactory performance of the proposed forensic framework in terms of statistical results, 2D and 3D views of position estimations.
Development of CAD System for Detection, Classification, Retrieval and 3d Reconstruction of Brain and Liver Tumors on MRI and CT Images
(National Institute of Technology Karnataka, Surathkal, 2014) Arakeri, Megha P.; Reddy, G. Ram Mohana
Brain and liver tumors are the life-threatening diseases due to low survival rate. Hence, accurate diagnosis of brain and liver tumors is necessary to provide effective treatment. Medical imaging techniques like Computed Tomography (CT) and Magnetic Resonance Imaging (MRI) help in acquiring images of the tumor. The visual analysis of these medical images by the radiologist is time consuming, subjective and inaccurate. The needle biopsy of the tumor provides accurate diagnosis but it is an invasive technique and generally not recommended. In order to overcome these drawbacks, there is a need to develop Computer-Aided Diagnosis (CAD) system for assisting the radiologist in fast and accurate diagnosis of tumors. Therefore, this thesis proposes an effective and efficient CAD system for tumor detection, classification, Content-Based Image Retrieval (CBIR), and 3D reconstruction to provide complete assistance to the radiologist in the diagnosis of brain and liver tumors. In the first methodology, this thesis aims at tumor detection, by proposing automatic, effective and efficient segmentation methods for brain and liver tumors on medical images. The brain tumor is detected using the proposed segmentation technique based on Modified Fuzzy C-Means (MFCM) clustering algorithm. The liver tumor is detected using the proposed segmentation technique based on the automatic region growing algorithm. In the second methodology, this thesis targets at identification of the type of brain/liver tumor as benign or malignant, by proposing an effective and efficient tumor classification scheme. Precisely, the proposed scheme represents the tumor characteristics using its significant features, selects most discriminating features using a two-level feature selection technique consisting of Information Gain (IG) based feature ranking and Independent Component Analysis (ICA) based feature section methods. Then, the tumor is classified using an ensemble classifier consisting of Support Vector Machine (SVM), Artificial Neural Network (ANN) and k-Nearest Neighbor (k-NN) classifiers. In the third methodology, this thesis proposes two CBIR methods based on image rotation correction and rotation invariant features to assist the radiologist in brain/liver tumor diagnosis based on past resolved cases. In order to provide fast retrieval of tumor images from the database, the tumor features in the database are indexed using the proposed indexing technique called as Cluster with IG-ICA and KD-tree (CIKD). The features in the database are partitioned into different groupsusing modified k-means clustering which identifies the number of clusters and initial cluster centers automatically. In the fourth methodology, this thesis aims to build the 3D model of the brain/liver tumor, by proposing an effective and efficient 3D reconstruction scheme. Precisely, it proposes an enhanced shape-based interpolation algorithm to estimate missing slices in a given set of brain/liver tumor slices. Further, the 3D mesh simplification algorithm is proposed to reduce the number of triangles in the reconstructed mesh and accelerate the rendering phase. The tumor volume is also computed to assist the radiologist in estimating the stage of cancer. Experiments are carried out on a dataset consisting of MRI images of the brain tumor and CT images of the liver tumor. Experimental results demonstrate that the proposed CAD system is automatic, effective and efficient in the diagnosis of brain and liver tumors.
Energy Efficient Resource Management and Task Scheduling at the Cloud Data Center
(National Institute of Technology Karnataka, Surathkal, 2018) Sharma, Neeraj Kumar; Reddy, G. Ram Mohana
Due to the growing demand for cloud services, allocation of energy efficient resources (CPU, memory, storage, etc.) and utilization of these resources are the major challenging issues of a large cloud data center. To meet the ever increasing demand of the customers, more number of servers are needed at the data center. These data centers require more cooling devices in order to keep the data center at a specified temperature resulting in more energy consumption and CO2 emission. The user requested on demand virtual machine (VM) allocation problem is widely known as a combinatorial optimization problem. Due to the large number of PMs present in the data center, the specified VM allocation problem is related to the NP-hard/NP-complete complexity class. Finding an optimal solution to the specified VM allocation problem with the multi-objective approach in the polynomial time will thus create a lot of challenges. Further, the networking devices of data center like switches consume 10% to 20% of the total energy consumed by IT devices in the data center. Hence, the network-aware VM allocation algorithm is required to minimize the energy consumption of switches and physical machines (PMs) at the cloud data center. Further, a policy for migrating VMs from underutilized PMs to the energy efficient PMs is required over a period of time without violating the service level agreement (SLA) between the cloud service provider and the customer. In order to minimize both the energy consumption and resources wastage, this thesis presents multi-objective VM allocation to PM using hybrid bio-inspired algorithms (HGACSO, HGAPSO, and HGAPSOSA) based on GA, CSO, PSO, and SA algorithms. Further, to save the energy consumption of networking switches in the cloud data center, a branch-and-bound based exact algorithm is proposed for VM allocation problem. The proposed branch-and-bound based exact algorithm saves the energy consumption of PMs and networking switches at the cloud data center. Further, the proposed VM migration technique and a task scheduling technique based on the First-Fit approximation algorithm will not only reduce the energy consumption at the cloud data center but also avoids the SLA. The experimental results were carried out in both homogeneous and heterogeneous cloud data center environments. Experimental results demonstrated that the proposed VM allocation algorithms outperform the state-of-the-art benchmark and peer research algorithms.
Optimizing Vertical Handover Decision Making in Heterogeneous Wireless Networks
(National Institute of Technology Karnataka, Surathkal, 2016) Chandavarkar, Beerappa Rama; Reddy, G. Ram Mohana
Ever-increasing demands of users and the development of modern communication technologies have led to the evolution of 4th Generation (4G) heterogeneous wireless networks. The integration of wireless networks of different characteristics and the demands of: user, mobile device, applications and service providers result in issues such as seamless mobility management, security, administration, billing, etc. The key issue among these challenges is the handover process of mobility management for seamless communication of mobile devices in heterogeneous wireless networks with maximized users’ satisfaction. Always Best Connected (ABC) services anywhere at anytime is one of the key objectives of 4G in integrating IEEE and cellular technologies. This thesis mainly addresses the Vertical Handover Decision (VHD) making in heterogeneous wireless networks for seamless communication of mobile devices. The dependency of VHD on multiple attributes in heterogeneous wireless networks demands an optimized handover process in terms of minimized complexity with improved reliability and flexibility. Several existing methods like fuzzy logic, neural networks, game theory and Multiple Attribute Decision Making (MADM) have been used for VHD. However there are still open issues such as, complexity, reliability and flexibility in these methods. MADM is one such method which supports multiple attributes based decision with minimum complexity for multiple criteria dependent VHD in heterogeneous wireless networks. The main problem with the MADM method is unreliable network selection and the rank reversal problem due to its dependency on attributes normalization and weight calculation methods. Hence, this thesis presents an optimized MADM method referred to as Simplified and Improved Multiple Attributes Alternate Ranking (SI-MAAR) for overcoming the limitations of classical MADM methods. Thus, SI-MAAR method is optimized in terms of minimized computational complexity and improved network selection reliability with the elimination of rank reversal problem. With MATLAB simulations, the analytical model of SI-MAAR method is demonstrated for 100% reliable VHD with the 0% rank reversal problem in heterogeneous wireless networks. iFurther, many of the classical MADM methods used in VHD in heterogeneous wireless networks depend on attributes weight computation techniques such as, Entropy, Variance, Analytical Hierarchy Process (AHP) etc. Expectations of users and applications during VHD in heterogeneous wireless networks is subjective in nature. AHP is one such popular method which supports computation of subjective attributes weight. The main problem with AHP is computation of reciprocal matrix through the involvement of the decision maker which will result in unreliable attributes weight and further to unreliable network selection in VHD. Hence, this thesis also presents an optimized AHP method referred to as Simplified and Improved Analytical Hierarchy Process (SI-AHP) to overcome unreliability in attributes weight computation. Thus, SI-AHP method is optimized in terms of minimum involvement of the decision maker resulting in reduced attributes weight computational complexity with the improved reliability. With MATLAB simulations, SI-AHP method is demonstrated for 100% reliable attributes weight computation used for VHD in heterogeneous wireless networks. In this thesis, SI-MAAR and SI-AHP methods are numerically analysed using MATLAB simulations and results demonstrate that SI-MAAR and SI-AHP methods are outperforming classical MADM and AHP methods respectively. Similarly, simulations using network simulators and further validation by testbed-based approaches are also required for justifying the proposed analytical solutions. Among the available open source network simulators such as NS2, NS3, OMNET++ and J-Sim, simulation of heterogeneous wireless networks is supported only in NS2’s distribution provided by National Institute of Science and Technology (NIST). The major problems with the NIST’s NS2 distribution are: (i) support for only one mobile node simulations (ii) minimal support for VHD and (iii) non-availability of configuration and result analysis tools such as TCP Performance Evaluation suite for simulations of heterogeneous wireless networks. Hence, this thesis also presents NS2 based Evaluation Suite for User Datagram Protocol applications referred to as “ES-UDP” for configuration, simulations and results analysis of multiple mobile nodes in heterogeneous wireless networks. Thus, ES-UDP tool provides both text and graphical results of handover, packets sent and received, throughput, packet delay and jitter of heterogeneous wireless networks simulations. iiOn the other hand, real time experimentation is subject to testbed’s deployment complexity, cost and time. The other major challenge of testbed experimentations is Linux kernel support for heterogeneous wireless networks. Although for handover execution, network layer protocol-Mobile IPv6 is supported by Linux kernel, the major issues are lack of testbed deployment informations, high cost, and nonavailability of testing checkpoints and debugging procedures. Thus, this thesis also presents a cost-effective testbed of Mobile IPv6 for handover execution in homogeneous and heterogeneous wireless networks. Further, this testbed can be used for VHD by deploying proposed solutions: SI-MAAR and SI-AHP in the Linux kernel. To summarize, the main contributions of the thesis are, improving the network selection reliability and the elimination of rank reversal problem of classical MADM methods used in VHD of heterogeneous wireless networks with “SI-MAAR”. Simplifying the attributes weight computation and improving the attributes weight reliability of AHP with “SI-AHP”. Ease of NS2’s configuration, simulation and results analysis of multiple mobile nodes experimentations in heterogeneous wireless networks with “ES-UDP” tool. Finally, simple and cost-effective Mobile IPv6 testbed for handover execution and further to handover decision in heterogeneous wireless networks
Resource Consumption Analysis of Virtualised Server Consolidation System
(National Institute of Technology Karnataka, Surathkal, 2018) Mohan, Biju R; Reddy, G. Ram Mohana
Resource consumption analysis is necessary because of continuous performance degradation of any long-running computing systems. Performance degradation is due to operating system's resource shrinkage. The most common causes of performance degradation include memory resource leakages, unreleased le descriptors, and numerical approximation errors. It is observed from literature that memory exhaustion has contributed majorly to the system failure. Resource consumption analysis is essential in a virtualized server consolidation system because Virtual Machines (VMs) use resources on demand. Another reason for selecting virtualized server consolidation system is due to the increased popularity of cloud computing. The key motivation behind this work is to help the system administrators to avoid accidental outage due to resource crunch. The key challenges in analyzing resource consumption data in server virtualized system are the volatility of the data and structural changes in the data. First, this thesis focussed on establishing performance degradation/aging e ect in virtualized server consolidation system. Then, we studied the e ectiveness of ARIMA models for forecasting the resource consumption data of virtualized server consolidation system; we found the presence of heteroscedasticity in the residuals of ARIMA model. The presence of heteroscedasticity in the residuals motivated us to try heteroscedastic models like ARCH and GARCH for resource forecasting. Another hybrid model namely ARIMA-ANN is also tried for resource forecasting. By combining di erent models, it is possible to capture various aspects of the underlying patterns. But we have experienced a slackness of t in all these models namely ARIMA, ARIMA-ARCH, ARIMA-GARCH, and ARIMA-ANN for the considered data. This slackness of t is due to the presence of structural changes in the resource consumption data. Further, Regime-Switching Models like MS-GARCH and SETAR are also used to analyze the data and found that these models have reasonably tted the considered data very well. Since there is no clear strategy for nding the order of GARCH and ARCH models, hence we tried di erent models and thus selected one model with least AIC, BIC, and log likelihood values for resource forecasting. An interested statistician could further investigate other mechanisms for nding the order of ARCH and GARCH models. As an extension, we would like to try these models and study the reasons for software aging in mobile platforms like Android systems in the near future.