P. Venkata Krishna

Honey bee behavior inspired load balancing of tasks in cloud computing environments

Scheduling of tasks in cloud computing is an NP-hard optimization problem. Load balancing of non-preemptive independent tasks on virtual machines (VMs) is an important aspect of task scheduling in clouds. Whenever certain VMs are overloaded and remaining VMs are under loaded with tasks for processing, the load has to be balanced to achieve optimal machine utilization. In this paper, we propose an algorithm named honey bee behavior inspired load balancing (HBB-LB), which aims to achieve well balanced load across virtual machines for maximizing the throughput. The proposed algorithm also balances the priorities of tasks on the machines in such a way that the amount of waiting time of the tasks in the queue is minimal. We have compared the proposed algorithm with existing load balancing and scheduling algorithms. The experimental results show that the algorithm is effective when compared with existing algorithms. Our approach illustrates that there is a significant improvement in average execution time and reduction in waiting time of tasks on queue.

Learning Automata-Based QoS Framework for Cloud IaaS

This paper presents a Learning Automata (LA)-based QoS (LAQ) framework capable of addressing some of the challenges and demands of various cloud applications. The proposed LAQ framework ensures that the computing resources are used in an efficient manner and are not over- or under-utilized by the consumer applications. Service provisioning can only be guaranteed by continuously monitoring the resource and quantifying various QoS metrics, so that services can be delivered in an on-demand basis with certain levels of guarantee. The proposed framework helps in ensuring guarantees with these metrics in order to provide QoS-enabled cloud services. The performance of the proposed system is evaluated with and without LA, and it is shown that the LA-based solution improves the performance of the system in terms of response time and speed up.

A learning automata-based fault-tolerant routing algorithm for mobile ad hoc networks

Reliable routing of packets in a Mobile Ad Hoc Network (MANET) has always been a major concern. The open medium and the susceptibility of the nodes of being fault-prone make the design of protocols for these networks a challenging task. The faults in these networks, which occur either due to the failure of nodes or due to reorganization, can eventuate to packet loss. Such losses degrade the performance of the routing protocols running on them. In this paper, we propose a routing algorithm, named as learning automata based fault-tolerant routing algorithm (LAFTRA), which is capable of routing in the presence of faulty nodes in MANETs using multipath routing. We have used the theory of Learning Automata (LA) for optimizing the selection of paths, reducing the overhead in the network, and for learning about the faulty nodes present in the network. The proposed algorithm can be juxtaposed to any existing routing protocol in a MANET. The results of simulation of our protocol using network simulator 2 (ns-2) shows the increase in packet delivery ratio and decrease in overhead compared to the existing protocols. The proposed protocol gains an edge over FTAR, E2FT by nearly 2% and by more than 10% when compared with AODV in terms of packet delivery ratio with nearly 30% faulty nodes in the network. The overhead generated by our protocol is lesser by 1% as compared to FTAR and by nearly 17% as compared to E2FT when there are nearly 30% faulty nodes.

An adaptive learning routing protocol for the prevention of distributed denial of service attacks in wireless mesh networks

Wireless Mesh Networks (WMNs) have potentially unlimited applications in the future. Therefore, establishing a viable and secure wireless network routing protocol for these networks is essential. Currently, these networks are being used in connecting large sections of cities by setting up wireless routers at strategic points all around the city. These networks can also support connecting remote areas of the country, instead of having to lay a cable all the way. The nature of applications mentioned above make these networks prone to different attacks. Thus, security of these networks is a serious concern. In this paper, we study the impact of Distributed Denial of Service (DDoS) attacks on WMNs. We base our work on the existing Optimized Link State Routing protocol (OLSR) and we weave in concepts of Learning Automata (LA) to protect the network from this kind of attack. The simulation results for the proposed scheme show that the proposed protocol is effective in the prevention of DDoS attacks in WMNs.

LACAV: an energy-efficient channel assignment mechanism for vehicular ad hoc networks

Designing an efficient channel assignment system for Vehicular Ad hoc Networks (VANETs), which conserves energy, is a challenging task, primarily because of the high degrees of mobility of nodes in these networks. As the high mobility of nodes in vehicular networks leads to frequent handoffs, channel assignment in VANETs becomes a tedious task. In this paper, we propose a channel assignment mechanism using the concepts of learning automata (LA) and reusability. LA is used to optimize the performance of the proposed system by selecting suitable number of reserved channels for the handoff calls and reusability allows the channel to be reused by the different base stations (BSs) based on the reuse distance. The proposed system is designed to reduce the dropping probability. The proposed system is suitable for network architectures in which it is possible to arrange the BSs with different groups of channels sequentially in a particular order that helps in conserving energy. Our experiments clearly indicate that the system reduces the dropping probability and allows a continuous communication throughout the duration of the call. The performance of proposed algorithm is compared with the Vehicular Fast Handover Scheme (VFHS), and the Cooperative scheme for service channel reservation (CRaSCH) scheme in terms of handoff latency, and it is shown that the proposed algorithm performs better than VFHS and CRaSCH.

Quality-of-service-enabled ant colony-based multipath routing for mobile ad hoc networks

Mobile ad hoc networks (MANETs) are dynamically changing and self-configuring networks. Owing to their widespread use for many applications, multipath routing in MANETs has been widely discussed for providing fault-tolerance routing, quality-of-service (QoS) and various other purposes. The authors propose a quality of service enabled ant colony-based multipath routing (QAMR) algorithm based on the foraging behaviour of ant colony for selecting path and transmitting data. In this approach, the path is selected based on the stability of the nodes and the path preference probability. The authors have considered bandwidth, delay and hop count as the QoS parameters along with the stability of node, number of hops and path preference probability factors. Simulations performed with network simulator 2 shows that the proposed algorithm is scalable and performs better at higher traffic load compared to the existing algorithms.

LAID: a learning automata-based scheme for intrusion detection in wireless sensor networks

In this paper, we address the problem of intrusion detection in wireless sensor networks (WSNs) using a learning automata (LA)-based approach. We are not aware of any LA-based intrusion detection systems (IDSs) for WSN. Additionally, the S-model approach that we have taken to solve the problem, wherein the feedback of the environment to the automaton can not only be completely favorable or completely unfavorable, but also be any continuous value within these extremities, makes it one of the attractive solution approaches in LA. We have rigorously evaluated the performance of our proposed solution by performing a variety of experiments and have found our solution approach to be promising. Copyright

An efficient approach for distributed dynamic channel allocation with queues for real-time and non-real-time traffic in cellular networks

We are witnessing these days a rapid growth of mobile users. Therefore, frequency spectrum must be efficiently utilized, as available frequency spectrum is limited. This paper proposes a channel allocation scheme with efficient bandwidth reservation, which initially reserves some channels for handoff calls, and later reserves the channels dynamically, based on the user mobility. The direction of user mobility may not be straight always, but the user may also go left, right or backwards. Thus, QoS can be improved, if the channel reservation is made based upon the user mobility and the location of the user. We devise here a new algorithm that deals with multiple traffic systems by modifying the existing DDCA algorithm [Krishna, P.V., Iyengar, N.Ch.S.N., 2008. Optimal channel allocation algorithm with efficient channel reservation for cellular networks. International Journal of Communication Networks and Distributed Systems 1 (1), 33-51]. This algorithm reserves more channels for hot cells, less number of channels for cold cells and an average number of channels for the medium cells. Furthermore, we maintain queues for all types of calls. We model the system by a three-dimensional Markov Chain and compute the QoS parameters in terms of the blocking probability of originating calls and the dropping probability of handoff calls. The results indicate that the proposed channel allocation scheme exhibits better performance by considering the above mentioned user mobility, type of cells, and maintaining of the queues for various traffic sources. In addition, it can be observed that our approach reduces the dropping probability by using reservation factor.

Learning Automata Based Sentiment Analysis for recommender system on cloud

The development of personalized recommendation systems has been an interesting research topic after the rapid evolution in social networking sites. In this paper, we propose a recommendation system using Learning automata (LA) and sentiment analysis. LA is used to optimize the recommendation score produced by the proposed system using sentiment analysis. The proposed Learning Automata-Based Sentiment Analysis System (LASA) recommends the places nearby the current location of the users by analyzing the feedback from the places and thus calculating the score based on it. Experiments performed by us indicate that by using LA, we can improve the performance of the proposed system, and, thus, help a user to find a specific location according to the need.

Versatile time-cost algorithm VTCA for scheduling non-preemptive tasks of time critical workflows in cloud computing systems

In cloud computing environments, resources and infrastructure are provided as a service over internet on demand. The users are interested in reducing the service cost provided by the cloud service providers. Scheduling tasks of workflows play a vital role in determining performance of cloud computing systems. Workflows have many tasks in it and are interdependent on each other. Time critical workflows comprise of a collection of tasks which should be completed as early as possible so that other workflows get its turn. The budget involved in executing the time critical tasks is very high. The execution cost increases whenever we try to reduce the execution time. In this paper, we propose a method called versatile time-cost algorithm VTCA to schedule time critical workflows with minimum cost. VTCA will schedule the tasks to complete in earliest possible time as well as optimise the cost involved in resource provisioning. The results of experiments conducted using CloudSim simulator show that our scheduling policy minimises the completion time of workflows than other existing algorithms like min-min and fair max-min by 5% to 30% and it also reduces the costs by 5% to 35%.