Schedulers for cloud computing determine on which processing resource jobs of a workflow should be allocated. In hybrid clouds, can allocated either private or public pay per use basis. The capacity the communication channels connecting these two types resources impacts makespan and cost execution. This article introduces scheduling problem in clouds presenting main characteristics to considered when workflows, as well brief survey some algorithms used systems. To assess influence job...

Traditional routing protocols employ limited information to make decisions, which can lead a slow adaptation traffic variability, as well restricted support the Quality of Service (QoS) requirements applications. This article introduces novel approach for in Software-defined networking (SDN), called Reinforcement Learning and Software-Defined Networking Intelligent Routing (RSIR). RSIR adds Knowledge Plane SDN defines algorithm based on (RL) that takes into account link-state decisions....

Traditional routing protocols employ limited information to make decisions, which leads slow adaptation traffic variability and restricted support the quality of service requirements applications. To address these shortcomings, in previous work, we proposed RSIR, a solution based on Reinforcement Learning (RL) Software-Defined Networking (SDN). However, RL-based solutions usually suffer an increase time during learning process when dealing with large action state spaces. This paper...

This article presents a method for the computation of equivalent bandwidth an aggregate heterogeneous self-similar sources, as well time scales interest queueing systems fed by fractal Brownian motion (fBm) process. Moreover, leaky bucket, novel policing mechanism capable accurately monitoring is introduced.

Network connectivity is a key issue in the realization of IoT, and LTE cellular technology most promising option for provisioning such connectivity. However, networks, large number IoT devices trying to access medium can overload RAN. In this article, we review random procedure its support applications. We also assess performance RAN control schemes proposed by 3GPP, taking into consideration interaction between packet downlink channel resource allocation.

The Internet of Things (IoT) paradigm stands for virtually interconnected objects that are identifiable and equipped with sensing, computing, communication capabilities. Services applications over the IoT architecture can take benefit long-term evolution (LTE)/LTE-Advanced (LTE-A), cellular networks to support machine-type (MTC). Moreover, it is paramount MTC do not affect services provided traditional human-type (HTC). Although previous studies have evaluated impact number devices on...

In the fog computing paradigm, nodes are placed on network edge to meet end-user demands with low latency, providing possibility of new applications. Although role cloud remains unchanged, a infrastructure for must be created. The design such an consider user mobility, which causes variations in workload demand over time different regions. Properly deciding location is important reduce costs associated their deployment and maintenance. To these demands, this paper discusses problem locating...

The support of stringent requirements such as ultra-low latency and ultra-reliability the forthcoming 5G services poses several challenges to telecommunications infrastructure providers. Network Function Virtualization, multi-access edge computing (MEC), network slicing capabilities can help requirements. However, a trade-off between cost resource deployment service needs be taken into account in design NFV-based networks. In this paper, we investigate MEC location problem, which aims at...

Network slicing is a promising technology for providing customized logical and virtualized networks the fifth-generation (5G) use-cases (enhanced mobile broadband, ultrareliable low-latency communications, massive machine-type communications), which pose distinct quality of service (QoS) requirements. Admission control resource allocation mechanisms are pivotal realizing network efficiently, but existing focus on radio access network. This article proposes an approach encompassing...

In hybrid clouds, inter-cloud links play a key role in the execution of jobs with data dependencies. Insufficient available bandwidth can increase makespan and monetary cost to execute application on public clouds. Imprecise information about lead inefficient scheduling decisions. This paper attempts evaluate impact imprecise workflow schedules, it proposes mechanism cope its estimates. The proposed applies deflating factor value furnished as input scheduler. Simulation results showed that...

Fog computing brings the cloud close to users, reducing latency and allowing deployment of new delay sensitive applications. Fogs clouds can work cooperatively improve service delivery end users. An essential aspect a fog-cloud system is decision-making process on where allocate resources run tasks an application. This paper introduces novel mechanism named Gaussian Process Regression for Fog-Cloud Allocation (GPRFCA) resource allocation in infrastructure composed cooperative fogs clouds....

In flex-grid (elastic) networks, the spectrum can be allocated at a much finer granularity than it in WDM networks. However, dynamic establishment and tear down of lightpaths yields to fragmentation with consequent increase blocking requests for connection establishment. Therefore, is paramount importance that allocation decisions try mitigate problem. line that, this paper introduces Multigraph Shortest Path Algorithm novel Routing Spectrum Allocation (RSA) elastic Results indicate joint...

This paper investigates the problem of locating edge devices in ultra-dense 5G network function virtualization-based small cell networks for provisioning reliable broadcasting services. The is formulated as a capacitated facility location (CRFLP) with failure probability. solution aims at placing VNFs broadcast transmissions optimally on selected order to ensure high reliability and minimize cost providing services well probability loss service requests. CRFLP non-linear NP-hard it evaluated...

Space division multiplexing is a promising solution, which has been proposed for elastic optical networks to deal with the anticipated exhaustion of capacity single core networks. This letter introduces PERFECTA algorithm, novel algorithm provide failure-independent path protecting p-cycle protection in using modulation and space multiplexing.

This paper introduces a mechanism which dynamically tunes the parameters of 802.11e contention-based access method. The proposed aims at providing QoS as well ameliorating problem delay asymmetry

Network virtualization is a promising technique for building the Internet of future since it enables low cost introduction new features into network elements. An open issue in such how to effect an efficient mapping virtual elements onto those existing physical network, also called substrate network. Mapping NP-hard problem and solutions ignore various real characteristics order solve reasonable time frame. This paper introduces algorithms this based on 0–1 integer linear programming, whole...

Mobile Operators have experienced a growth of demand new services with strict requirements and features which imposes challenges to operate the mobile wireless network. To tackle these challenges, Telcos must provide flexible, dynamic network architecture, furnishing resiliency ensure end-to-end (E2E) service continuity in case resource failures. Virtualization core elements cellular Long-Term Evolution (LTE) LTE evolved Multimedia Broadcast Multicast Service (eMBMS) technology has been...

The processing of big data generated by the Industrial Internet Things (IIoT) calls for support at edge network, as well cloud centers. equal-cost multipath, which is default routing technique in centers, can degrade network performance when handling mouse and elephant flows. Such degradation compromise strict quality service requirements IIoT over 5G networks. Novel techniques scheduling flows alleviate this problem. Recently, several approaches have incorporated machine learning...