1. Faculty Publications
Permanent URI for this communityhttps://idr.nitk.ac.in/handle/1/5
Browse
6 results
Search Results
Item Stochastic dynamic programming model for optimal resource allocation in vehicular ad hoc networks(2018) Bhuvaneswari, M.; Paramasivan, B.; Kandasamy, A.Vehicular ad hoc network (VANET) is an emerging trend where vehicles communicate with each other and possibly with a roadside unit to assist various applications like monitoring, managing and optimizing the transportation system. Collaboration among vehicles is significant in VANET. Resource constraint is one of the great challenges of VANETs. Because of the absence of centralized management, there is pitfall in optimal resource allocation, which leads to ineffective routing. Effective reliable routing is quite essential to achieve intelligent transportation. Stochastic dynamic programming is currently employed as a tool to analyse, develop and solve network resource constraint and allocation issues of resources in VANET. We have considered this work as a geographical-angular-zone-based two-phase dynamic resource allocation problem with a homogeneous resource class. This work uses a stochastic dynamic programming algorithm based on relaxed approximation to generate optimal resource allocation strategies over time in response to past task completion status history. The second phase resource allocation uses the observed outcome of the first phase task completion to provide optimal viability in resulting decisions. The proposed work will be further extended for the scenario that deals with heterogeneous resource class. Simulation results show that the proposed scheme works significantly well for the problems with identical resources. 2018, Indian Academy of Sciences.Item Elucidating the challenges for the praxis of fog computing: An aspect-based study(2019) Martin, J.P.; Kandasamy, A.; Chandrasekaran, K.; Joseph, C.T.The evolutionary advancements in the field of technology have led to the instigation of cloud computing. The Internet of Things paradigm stimulated the extensive use of sensors distributed across the network edges. The cloud datacenters are assigned the responsibility for processing the collected sensor data. Recently, fog computing was conceptuated as a solution for the overwhelmed narrow bandwidth. The fog acts as a complementary layer that interplays with the cloud and edge computing layers, for processing the data streams. The fog paradigm, as any distributed paradigm, has its set of inherent challenges. The fog environment necessitates the development of management platforms that effectuates the orchestration of fog entities. Owing to the plenitude of research efforts directed toward these issues in a relatively young field, there is a need to organize the different research works. In this study, we provide a compendious review of the research approaches in the domain, with special emphasis on the approaches for orchestration and propose a multilevel taxonomy to classify the existing research. The study also highlights the application realms of fog computing and delineates the open research challenges in the domain. 2019 John Wiley & Sons, Ltd.Item Exploring the support for high performance applications in the container runtime environment(2018) Martin, J.P.; Kandasamy, A.; Chandrasekaran, K.Cloud computing is the driving power behind the current technological era. Virtualization is rightly referred to as the backbone of cloud computing. Impacts of virtualization employed in high performance computing (HPC) has been much reviewed by researchers. The overhead in the virtualization layer was one of the reasons which hindered its application in the HPC environment. Recent developments in virtualization, especially the OS container based virtualization provides a solution that employs a lightweight virtualization layer and promises lesser overhead. Containers are advantageous over virtual machines in terms of performance overhead which is a major concern in the case of both data intensive applications and compute intensive applications. Currently, several industries have adopted container technologies such as Docker. While Docker is widely used, it has certain pitfalls such as security issues. The recently introduced CoreOS Rkt container technology overcomes these shortcomings of Docker. There has not been much research on how the Rkt environment is suited for high performance applications. The differences in the stack of the Rkt containers suggest better support for high performance applications. High performance applications consist of CPU-intensive and data-intensive applications. The High Performance Linpack Library and the Graph500 are the commonly used computation intensive and data-intensive benchmark applications respectively. In this work, we explore the feasibility of this inter-operable Rkt container in high performance applications by running the HPL and Graph500 applications and compare its performance with the commonly used container technologies such as LXC and Docker containers. 2018, The Author(s).Item Fuzzy Reinforcement Learning based Microservice Allocation in Cloud Computing Environments(2019) Joseph, C.T.; Martin, J.P.; Chandrasekaran, K.; Kandasamy, A.Nowadays the Cloud Computing paradigm has become the defacto platform for deploying and managing user applications. Monolithic Cloud applications pose several challenges in terms of scalability and flexibility. Hence, Cloud applications are designed as microservices. Application scheduling and energy efficiency are key concerns in Cloud computing research. Allocating the microservice containers to the hosts in the datacenter is an NP-hard problem. There is a need for efficient allocation strategies to determine the placement of the microservice containers in Cloud datacenters to minimize Service Level Agreement violations and energy consumption. In this paper, we design a Reinforcement Learning-based Microservice Allocation (RL-MA) approach. The approach is implemented in the ContainerCloudSim simulator. The evaluation is conducted using the real-world Google cluster trace. Results indicate that the proposed method reduces both the SLA violation and energy consumption when compared to the existing policies. � 2019 IEEE.Item Comparative study of simulation tools and challenging issues in cloud computing(2018) Shishira, S.R.; Kandasamy, A.; Chandrasekaran, K.Resource Scheduling lays a key role in large-scale cloud applications. It is difficult for the developers to do an extensive research on all the issues in real time as it requires infrastructure which is beyond the control, also network condition cannot be predicted. Hence simulations are used which imitates the real time environment. There are various simulators developed for the research as it is difficult to maintain the infrastructure on premise. Thus to understand the tools in deep, we focused on five open source tools such as Cloudsim, CloudAnalyst, iCancloud, Greencloud and CloudSched. The above mentioned tools are compared based on the respective architecture, the process of simulation, structural elements and performance parameters. In the paper, we have also discussed some of the challenging issues among the tools for further research. � Springer Nature Singapore Pte Ltd. 2018.Item Toward efficient autonomic management of clouds: A CDS-based hierarchical approach(2018) Martin, J.P.; Kandasamy, A.; Chandrasekaran, K.Cloud computing is one of the most sought-after technologies today. Beyond a shadow of doubt, the number of clients opting for Cloud is increasing. This steers the complexity of the management of the Cloud computing environment. In order to serve the demands of customers, Cloud providers are resorting to more resources. Relying on a single managing element to coordinate the entire pool of resources is no more an efficient solution. Therefore, we propose to use a hierarchical approach for autonomic management. The problem that we consider here is to determine the nodes at which we have to place the Autonomic Managers (AMs), in order to ease the management process and minimize the cost of communication between the AMs. We propose a graph-theory-based model using Connected Dominating Set (CDS) that allows to determine an effective placement of AMs in different Data Centers (DCs), and, their collaboration with the Global Manager (GM). The approach considers the construction of domination sets and then, distributing the control of the dominees among the dominators. � 2018, Springer Nature Singapore Pte Ltd.