The fifth generation of mobile communications is anticipated to open up innovation opportunities for new industries such as vertical markets. However, these verticals originate myriad use cases with diverging requirements that future 5G networks have to efficiently support. Network slicing may be a natural solution to simultaneously accommodate over a common network infrastructure the wide range of services that vertical-specific use cases will demand. In this article, we present the network slicing concept, with a particular focus on its application to 5G systems. We start by summarizing the key aspects that enable the realization of so-called network slices. Then, we give a brief overview on the SDN architecture proposed by the ONF and show that it provides tools to support slicing. We argue that although such architecture paves the way for network slicing implementation, it lacks some essential capabilities that can be supplied by NFV. Hence, we analyze a proposal from the ETSI to incorporate the capabilities of SDN into the NFV architecture. Additionally, we present an example scenario that combines SDN and NFV technologies to address the realization of network slices. Finally, we summarize the open research issues with the purpose of motivating new advances in this field.

The fifth-generation mobile initiative, 5G, is a tremendous and collective effort to specify, standardize, design, manufacture, and deploy the next cellular network generation. 5G networks will support demanding services such as enhanced Mobile Broadband, Ultra-Reliable and Low Latency Communications and massive Machine-Type Communications, which will require data rates of tens of Gbps, latencies of few milliseconds and connection densities of millions of devices per square kilometer. This survey presents the most significant use cases expected for 5G including their corresponding scenarios and traffic models. First, the paper analyzes the characteristics and requirements for 5G communications, considering aspects such as traffic volume, network deployments, and main performance targets. Secondly, emphasizing the definition of performance evaluation criteria for 5G technologies, the paper reviews related proposals from principal standards development organizations and industry alliances. Finally, well-defined and significant 5G use cases are provided. As a result, these guidelines will help and ease the performance evaluation of current and future 5G innovations, as well as the dimensioning of 5G future deployments.

YouTube currently accounts for a significant percentage of the Internet's global traffic. Hence, understanding the characteristics of the YouTube traffic generation pattern can provide a significant advantage in predicting user video quality and in enhancing network design. In this paper, we present a characterisation of the traffic generated by YouTube when accessed from a regular PC. On the basis of this characterisation, a YouTube server traffic generation model is proposed, which, for example, can be easily implemented in simulation tools. The derived characterisation and model are based on experimental evaluations of traffic generated by the application layer of YouTube servers. A YouTube server commences the download with an initial burst and later throttles down the generation rate. If the available bandwidth is reduced (e.g. in the presence of network congestion), the server behaves as if the data excess that cannot be transmitted because of the reduced bandwidth were accumulated at a server's buffer, which is later drained if the bandwidth availability is recovered. As we will show, the video clip encoding rate plays a relevant role in determining the traffic generation rate, and therefore, a cumulative density function for the most viewed video clips will be presented. The proposed traffic generation model was implemented in a YouTube emulation server, and the generated synthetic traffic traces were compared with downloads from the original YouTube server. The results show that the relative error between downloads from the emulation server and the original server does not exceed 6% for the 90% of the considered videos. Copyright © 2012 John Wiley & Sons, Ltd.

Abstract-Network Function Virtualization is considered one of the key technologies for developing the future mobile networks. In this paper, we propose a theoretical framework to evaluate the performance of an LTE virtualized Mobility Management Entity (vMME) hosted in a data center. This theoretical framework consists of i) a queuing network to model the vMME in a data center, and ii) analytic expressions to estimate the overall mean system delay and the signaling workload to be processed by the vMME. We validate our mathematical model by simulation. One direct use of the proposed model is vMME dimensioning, i.e., to compute the number of vMME processing instances to provide a target system delay given the number of users in the system. Additionally, the paper includes a scalability analysis of the system. In our study we consider the billing model and a data center setup of Amazon Elastic Compute Cloud service, and estimate experimentally the processing time of MME processing instances for different LTE control procedures. For the considered setup, our results show that a vMME is scalable for signaling workloads up to 37000 LTE control procedures per second for a target mean system delay of 1 ms. The database performance assumed imposes this limit in the system scalability.

Network Function Virtualization (NFV) is considered one of the key technologies for the 5G mobile networks. In NFV, network functions are implemented in software components denominated Virtual Network Functions (VNFs) running on commodity hardware. In this paper, we propose an analytical model based on an open queuing network of G/G/m queues to model VNFs with several components, and chains of VNFs. Our model is flexible and generic enough to capture the behavior of such systems. We validate our model by simulation. Specifically, we validate it for an LTE virtualized Mobility Management Entity with a three-tiered architecture use case. We also compare our model with the estate of the art, in terms of computational complexity and estimation error. The results show that our model has a computational complexity similar to the method for analyzing Jackson's networks. Additionally, our model exhibits an estimation error, measured as the relative error for the estimation of the mean response time, approximately equal to 10%, whereas for the considered baseline systems it ranges roughly from 60% to 90%.Index Terms-5G, NFV, VNF, analytical model, queuing model.