Search

Services in today’s network produce a heavy amount of traffic, and all this traffic might have to pass through a Service Function Chain (SFC). The service provider must place these SFCs in the appropriate locations, and scale them as the request to the service increases. Appropriate placement and scaling of these SFCs will ensure Service Level Agreement (SLA) in terms of throughput, end-to-end delay, or successful serving of the requests. SFC placement and scaling are correlated tasks, however need to be modeled separately. For placing SFCs we require to consider server resources such as memory, CPU, bandwidth, and server locations. On the other hand for scaling, we need to consider resource utilization of VNFs as extra matrixes. Scaling in and out would require an additional cost and time to deploy new VNFs, hence it is important to predict VNF resource requirements in advance and prepare the resources for incoming traffic needs. In this paper we predicted an incoming resource demand using Recurrent Neural Network (RNN) based Gated Recurrent Unit (GRU) model. We predict the resource demands 2 min in advance and deploy the VNFs using the EdgeQ-Leaning model. We call this integrated algorithm as GRU-EdgeQL. To validate our approach we compared it with threshold and random algorithms. We implemented our algorithm in the OpenStack testbed. Our validation reveals that GRU based prediction and EdgeQ-Learning based placement does not only meet the Service Level Agreement (SLA) requirement by reducing the overall latency and SLA violation but also reduces the number of scaling operations thereby reducing the overall scaling cost.

Clustering plays a vital role in extending the lifespan and optimized direction of a wireless sensor network by integrating sensor nodes through clusters and choosing cluster heads (CHs) and non-cluster heads (NCHs). Cluster head aggregated data and non-cluster heads forward to the base station (BS). In this paper, we have introduced a new Dynamic Multipath Routing Protocol (DMPRP) for selections of cluster heads (CHs) and non-cluster heads (NCHs), which is ideally selected using M-PSO algorithm. After calculating the probabilities, the best selection of cluster heads has taken, and results have used to find the optimized shortest path using the Genetic Algorithm (GA). The GA algorithm uses an objective function consisting of a network to determine the optimal path.

In this, we provide the frontline Internet of Things (IoT), enabler of the newest oral communication class between humans and devices. A most crucial strength of the IoT concept is that the great have an impact on it is on manner via growing a change dimension to the world, pretty like what the online as soon as did. The paper pronounces the definitions of IoT, associated challenges, and discusses the usage of IoT in the various fields and day-to-day environments. The content material consists of the strengths and obstacles of purposes supported the IoT inactivity, transportation, healthcare, environment, and disaster.

Biggest challenge in deploying Service Function Chain (SFC) in the Edge Computing environment is the lack of resources at the edge. Hence while finding the optimum path for SFC deployment, the resource constraint environment should be observed and incorporated well in deployment scenarios. In this paper, we developed an environment aware adaptive Q-Learning algorithm to find an optimal SFC deployment path in edge computing environment. The available servers are divided into hierarchical network structure with local, neighbor, and datacenter servers to model an edge computing environment. The resource dynamics in the environment is modeled as a state transition probability. We compared the new algorithm with our base case algorithm that solely depends on Q-Learning and doesn’t incorporate the state transition probabilities. An intuitive reward function is designed to give maximum reward to complex deployment with minimum delays. We integrated our algorithm with physical testbeds using OpenStack and open source REST APIs. We evaluated SFC deployment on physical testbed using 42 different scenarios by measuring RTT.

Reinforcement learning (RL) has been used in various path finding applications including games, robotics and autonomous systems. Deploying Service Function Chain (SFC) with optimal path and resource utilization in edge computing environment is an important and challenging problem to solve in Software Defined Network (SDN) paradigm. In this paper we used RL based Q-Learning algorithm to find an optimal SFC deployment path in edge computing environment with limited computing and storage resources. To achieve this, our deployment scenario uses a hierarchical network structure with local, neighbor and datacenter servers. Our Q-Learning algorithm uses an intuitive reward function which does not only depend on the optimal path but also considers edge computing resource utilization and SFC length. We defined regret and empirical standard deviation as evaluation parameters. We evaluated our results by making 1200 test cases with varying SFC-length, edge resources and Virtual Network Function's (VNF) resource demand. The computation time of our algorithm varies between 0.03~0.6 seconds depending on the SFC length and resource requirement.

Over‐the‐top (OTT) services such as Netflix, Amazon Prime, and YouTube generate the most dominant form of traffic on the Internet today. There is increasingly high demand for resource intensive 3D contents, interactive media, 360 media, and user‐generated contents. As the amount of contents keep increasing in multiple folds, it is important to cache contents intelligently. Caching algorithm needs to exploit in‐network caching, community‐based pre‐caching, and a combined approach. Hence, we survey CDN‐based edge caching infrastructures including OpenConnect (Netflix) and Google Edge, followed by CCN based in‐network caching. We implement and compare four different approaches for caching contents including (1) in‐network caching, (2) edge caching, (3) community‐based in‐network caching, and (4) community‐based edge caching. We run our algorithms on adaptive network conditions with different topologies, cache size, content popularity, and request arrivals in and compared the delay for all these four approaches. We verify our model by calculating important performance parameters including hop count, redundancy, and hop count variances. Hopcount is an important performance parameter as it influences the processing, queuing, and transmission delays. We focus on determining if an in‐network caching approach is any better than edge caching. We reach several conclusions. First, in most of the scenarios, community‐based in‐network caching performs the best. Second, if the cache size is lesser than 30% of the total content size then community‐based edge caching is better for less popular contents. Finally, our statistical analysis also reveals that a community‐based edge caching mechanism is least affected by varying cache sizes and dynamic user behavior, which makes it a better choice for providing Service Level Agreement.

Cloud computing gives on-demand access to conveyed assets on paid premise. Everyone needs to utilize these administrations to decrease the expense of framework and support, in this way the load on cloud is expanding step by step. Balancing the load is a standout amongst the most critical issue that distributed computing is confronting today. The load should to be appropriated reasonably among every one of the hubs. Load balancing can reduce the energy utilization. Every one of these algorithms works in various ways and has a few favorable circumstances and restrictions. The most imperative for load balancing calculations is to consider the qualities like throughput, adaptation to internal failure, overhead, execution, and reaction time and asset usage. This paper primarily centers on the idea of load Balancing, writing review on load Balancing systems and diverse estimation parameters.

Observing the dynamic scenarios that change quickly with time is the prime utilization of remote sensor systems. This difference in conduct is contemplated to either certain outside variables or impediment of some framework structures itself in unpredicted causality. To adjust such conditions, machine learning procedures are esteemed to be valuable in disposing of the requirement for pointless update. Further, the method dependent on machine learning urges numerous viable answers for boost utilizations of asset and in this manner improves the life expectancy of the sensor arrange. In this paper, simply discussion over wireless sensor network, its characteristics, comparing with wireless ad hoc networks along with various application of wireless sensor network in different fields has been extended.

In this paper we have done a feasibility study to develop a simulation model for Content Centric Network. Our goal is to form communities based on popular clustering algorithm on the simulated CCN network. Furthermore we aim to do a guided content delivery on CCN based on the community preferences. Though community formation and a guided delivery of content is not a new field. However an experimental simulation of these concepts on CCN network is novel. Basically we want to utilize CCN network and its advantages to implement a next generation CDN.

The applications of Wireless Sensor Networks (WSNs) are growing at rapid pace and providing pervasive computing environments. Energy constraints is the most critical issue in sensor applications and that needs be optimized to prolong the life of resource constrained sensor network. Clustering is an efficient technique to group the sensor nodes of entire network into number of clusters to support high scal ability and provide better data aggregation by efficient utilization of limited resources of sensor nodes and that prolongs network lifetime. In this paper, some widely explored clustering algorithms in WSNs are discussed on several aspects and characteristics such as clustering timings, clustering attributes, convergence rate etc. The advantages and disadvantages of corresponding clustering algorithms are also explained with suitable examples. The paper finally concludes with discussion on the challenges of clustering in WSNs with mentioning the future research topics

Cloud computing has boomed its horizon with large pace as commercial infrastructure in the IT industries meeting the vast requirements of computing resources. There are several issues such as load balancing, virtual machine migration, automated service provisioning, algorithm complexity, etc., demanding to be resolved. Each of these issues needs load balancing to be resolved. This aims for distributing the unwanted dynamic workload between the nodes residing in cloud and this desires of every computing resource must be assigned on proficient and reasonable ground. Load balancing has become crucial for efficient performance in distributed environments. Cloud computing is an emerging technology demanding more services and better results. Thus load balancing for the cloud is very interesting and important research area. Many algorithms are proposed to provide efficient techniques for assigning the client's requests to available cloud nodes. This paper studies cloud computing along with research challenges in load balancing. Load balancing has been a major issue for cloud computing environment. Efficient load balancing scheme ensures efficient resource utilization by providing the resources to cloud on-demand of users’ basis. By implementing appropriate scheduling criteria load balancing may prioritize users. The aim of this study is to peep in various load balancing algorithms to address its challenges in variety of cloud environment. This study provides a perspective view of the latest approaches in load balancing that will certainly help the future researchers in this field

Node localization is one of the major issues of wireless sensor networks that determines the physical positions of nodes. The location data is useful to many communication protocols such as packet routing and sensing coverage in wireless sensor networks. It has been a daunting task to develop a practical algorithm for node localization in an environment where sensor nodes are with limited power, low cost etc. In this paper a novel idea to conserve energy required in a localization algorithm by integrating it with duty cycled MAC protocol in wireless sensor network has been modelled. To model the idea the RSSI localization method and a sensor SMAC protocol to regulate the control of energy wastages has been considered. Finally, every minute aspect of sensor network localization model including the topology, mobility, and channel and propagation schemes with scheduled transmission access, have been discussed in this paper.

The Internet Protocol Television (IPTV) subscription rate has increased steadily since the introduction of IPTV Video on Demand (VoD) services. We have developed a model to determine the optimum deployment strategy for delivering IPTV VoD network services from the IPTV service providers' perspective. The analysis technique in this paper helps us determine how to support a particular IPTV VoD deployment having a given number of customers that each must receive a contracted Quality of Experience (QoE). The main QoE measures considered are server waiting time, one-way minimum delay and access network bandwidth consumption. All these models are made considering heterogeneous network conditions. The QoE models will help service providers make network deployment decisions including the number of servers, distance of servers from a community and desirable access network bandwidth capacity. Copyright © 2011 John Wiley & Sons, Ltd.

Precise network topology information is required to perform management activities such as fault detection, root cause analysis, performance monitoring, and load balancing in enterprise networks. To accomplish these management tasks, both network discovery and connectivity information are essential. This paper examines various problems with the existing topology discovery mechanisms and proposes an SNMP-based approach to discover physical as well as logical topology. We present algorithms for identifying network device types and discovering connectivity among them. The connectivity of end host and management information base (MIB)-enabled devices with switches and routers is discussed and evaluated. We also present an algorithm for discovering logical topology, such as VLAN and subnet connectivity. Finally, we present a combination of graph and tree layouts, to visualize connectivity information.

This paper demonstrates the use of remote monitoring and processing to fine tune the measurements taken from the medical sensing devices. We demonstrated this with the help of use case related to glucose monitoring, however this method could be extended to other devices as well. Based on the technical requirement of signal transmission, the analysis for various communication protocols to provide these services is also investigated. A novel data mining technique is used to process the signals taken from the patient on the server side to improve the accuracy. Overall, the feasibility of the improved accuracy is experimented and demonstrated at the end. This paper exhibits the end-to-end solution for the given use case.

Machine to Machine (M2M) technology has the potential to increase the revenue, decrease the costs and improve the customer services of an organization. We have analyzed the management requirements of M2M systems, which are based on existing M2M network use cases and services. The most important characteristics including sleeping devices, low power lossy area networks, heterogeneous networks, device intelligence, mobility, two way communication, network dynamics, time sensitivity of data and data volume of M2M systems have been comprehensively investigated and reflected in management requirements discussed in this paper. The main management functionalities are fault, configuration, mobility, QoS and security management.

Selecting clothes and taking trials of them at the stores are always being time consuming job. It becomes, sometimes, difficult to find clothes of desired color, fabric and size. It is more difficult for customers who use e-shopping for their shopping as visualization is not possible. In this paper we are proposing a web-based system based on 3D human modeling and its visualization. System is deployed in two parts, one on client system and other on the server. Client system accepts customer's dimensions of their exposed body part, taken in traditional fashion, and 2D image of their face as input. Client system generates a 3D model using these inputs along with the face of the customer. Color of skin of the model is same as that of the 2D face image. Customers can customize their clothing style using digital styles and accessories present in the digital library at client system. After completion of clothing modification user can place orders. Client system sends all information, size and model, to the server.

The IPTV subscription rate has increased steadily since the introduction of IPTV Video on Demand (VoD) services. We have developed a model to determine the optimum deployment strategy for IPTV delivery network from the IPTV service providers' perspective. The analysis technique in this paper helps us determine the best deployment scenarios to support a certain number of customers within the tolerant boundary of Quality of Experience (QoE) measures. We define important QoE measures. The QoE will help service providers make deployment decisions including the number of servers, distance of servers from a community, and desirable access network bandwidth capacity.