Enabling IoT Network Slicing with Network Function Virtualization

Numerous Internet of Things (IoT) devices are being connected to the net-works to offer services. To cope with a large diversity and number of IoT ser-vices, operators must meet those needs with a more flexible and efficient net-work architecture. Network slicing in 5G promises a feasible solution for this issue with network virtualization and programmability enabled by NFV (Net-work Functions Virtualization). In this research, we use virtualized IoT plat-forms as the Virtual Network Functions (VNFs) and customize network slices enabled by NFV with different QoS to support various kinds of IoT services for their best performance. We construct three different slicing systems including: 1) a single slice system, 2) a multiple customized slices system and 3) a single but scalable network slice system to support IoT services. Our objective is to compare and evaluate these three systems in terms of their throughput, aver-age response time and CPU utilization in order to identify the best system de-sign. Validated with our experiments, the performance of the multiple slicing system is better than those of the single slice systems whether it is equipped with scalability or not.

Tackling IoT Scalability with 5G NFV-Enabled Network Slicing

With emerging large volume and diverse heterogeneity of Internet of Things (IoT) applications, the one-size-fits-all design of the current 4G networks is no longer adequate to serve various types of IoT applications. Consequently, the concepts of network slicing enabled by Network Function Virtualization (NFV) have been proposed in the upcoming 5G networks. 5G network slicing allows IoT applications of different QoS requirements to be served by different virtual networks. Moreover, these network slices are equipped with scalability that allows them to grow or shrink their instances of Virtual Network Functions (VNFs) when needed. However, all current research only focuses on scalability on a single network slice, which is the scalability at the VNF level only. Such a design will eventually reach the capacity limit of a single slice under stressful incoming traffic, and cause the breakdown of an IoT system. Therefore, we propose a new IoT scalability architecture in this research to provide scalability at the NS level and design a testbed to implement the proposed architecture in order to verify its effectiveness. For evaluation, three systems are compared for their throughput, response time, and CPU utilization under three different types of IoT traffic, including the single slice scaling system, the multiple slices scaling system and the hybrid scaling system where both single slicing and multiple slicing can be simultaneously applied. Due to the balanced tradeoff between slice scalability and resource availability, the hybrid scaling system turns out to perform the best in terms of throughput and response time with medium CPU utilization.

Evaluation of 5G Core Slicing on User Plane Function

Network slicing is one of the most important features in 5G which enables a large variety of services with diverse performance requirements by network virtualization. Traditionally, the network can be viewed as a one-size-fits-all slice and its services are bundled with proprietary hardware supported by telecom equipment providers. Now with the network virtualization technology in 5G, open networking software can be deployed flexibly on commodity hardware to offer a multi-slice network where each slice can offer a different set of network services. In this research, we propose a multi-slice 5G core architecture by provisioning its User Plane Functions (UPFs) with different QoS requirements. We compare the performance of such a multi-slice system with that of one-size-fits-all single slice architecture under the same resource assignment. Our research objective is to compare the performance of a network slicing architecture with that of a “one-size-fits-all” architecture and validate that the former can achieve better performance with the same underlying infrastructure. The results validate that our proposed system can achieve better performance by slicing one UPF into three with proper resource allocation.

The Roles of 5G Mobile Broadband in the Development of IoT, Big Data, Cloud and SDN

The fast technology development of 5G mobile broadband (5G), Internet of Things (IoT), Big Data Analytics (Big Data), Cloud Computing (Cloud) and Software Defined Networks (SDN) has made those technologies one after another and created strong interdependence among one another. For example, IoT applications that generate small data with large volume and fast velocity will need 5G with characteristics of high data rate and low latency to transmit such data faster and cheaper. On the other hand, those data also need Cloud to process and to store and furthermore, SDN to provide scalable network infrastructure to transport this large volume of data in an optimal way. This article explores the technical relationships among the development of IoT, Big Data, Cloud, and SDN in the coming 5G era and illustrates several ongoing programs and applications at National Chiao Tung University that are based on the converging of those technologies.

Coordinated Management of 5G Core Slices by MANO and OSS/BSS

To manage and orchestrate Network Slices (NSs) for 5G Core (5GC), the MANO (MANagement and Orchestration) framework is proposed by European Telecommunications Standard Institute (ETSI). In most research testbeds, MANO systems such as Tacker, OSM and ONAP are used to initiate network slices. However, this doesn’t comply with the 3GPP 5G standards as MANO should only be responsible for dynamic management of NSs, and the static management such as provisioning or unprovisioning a network slice should be left to OSS/BSS (Operation/Business Support System). Thus, in our testbed, an integrated architecture was designed in which the management of network slices will be coordinated by both MANO and OSS/BSS. MANO would handle on-boarding, instantiating, scaling and terminating of network slices while OSS/BSS is responsible for static management of slices including provisioning and unprovisioning of network slices. To evaluate our system, it was compared with the management systems equipped with only OSS/BSS or MANO in order to analyze the shortfalls of those systems when used to deploy network slices. Through this analysis, this research confirms the necessity of applying both OSS/BSS and MANO for the coordinated management of 5G core slices as adopted by 3GPP.

Enhancing Cloud-Based IoT/M2M System Scalability by Dynamic Network Slicing

With ever-increasing applications of IoT, and due to the heterogeneous and bursty nature of these applications, scalability has become an important research issue in building cloud-based IoT/M2M systems. This research proposes a dynamic SDN-based network slicing mechanism to tackle the scalability problems caused by such heterogeneity and fluctuation of IoT application requirements. The proposed method can automatically create a network slice on-the-fly for each new type of IoT application and adjust the QoS characteristics of the slice dynamically according to the changing requirements of an IoT application. Validated with extensive experiments, the proposed mechanism demonstrates better platform scalability when compared to a static slicing system.

Evaluating Dedicated Slices of Different Configurations in 5G Core

Network slicing is one of the most important concepts in 5G networks. It is enabled by the Network Function Virtualization (NFV) technology to allow a set of Virtual Network Functions (VNFs) to be interconnected to form a Network Service (NS). When network slices are created in 5G, some are shared among different 5G services while the others are dedicated to specific 5G services. The latter are called dedicated slices. Dedicated slices can be constructed with different configurations. In this research, dedicated slices of different configurations in 5G Core were evaluated in order to discover which one would perform better than the others. The performance of three systems would be compared: 1) Free5GC Stage 2 with each dedicated slice consisting of only UPF;2) Free5GC Stage 3 with each dedicated slice consisting of only UPF;3) Free5GC Stage 3 with each dedicated slice consisting of both SMF and UPF in terms of their registration time, response time, throughput, resource cost, and CPU utilization. It is shown that not one of the above systems will always be the best choice;based on the requirements, a specific system may be the best under a specific situation.