Adaptive Telemetry for Software-Defined Mobile Networks

Luca Cominardi, Sergio Gonzalez-Diaz, Antonio de la Oliva, Carlos J. Bernardos
2020 Journal of Network and Systems Management  
The forthcoming set of 5G standards will bring programmability and flexibility to levels never seen before. This has required introducing changes in the architecture of mobile networks, enabling different features such as the split of control and data planes, as required to support the rapid programming of heterogeneous data planes. Software Defined Networking (SDN) has emerged as a basic toolset for operators to manage their infrastructure, as it opens up the possibility of running a multitude
more » ... of intelligent and advanced applications for network optimization purposes in a centralized network controller. However, the very basic nature that makes possible this efficient management and operation in a flexible way-the logical centralizationposes important challenges due to the lack of proper monitoring tools, suited for SDN-based architectures. In order to take timely and right decisions while operating a network, centralized intelligence applications need to be fed with a continuous stream of up-to-date network statistics. However, this is not feasible with current SDN solutions due to scalability and accuracy issues. This article first analyzes the monitoring issues in current SDN solutions and then proposes a telemetry framework for software defined mobile networks capable of adapting to the various 5G services. Finally, it presents an experimental validation that shows the benefits of the proposed solution at alleviating the load on the control and data planes, improving the reactiveness to network events, and providing better accuracy for network measurements. • OAM defines stateful mechanisms that must be executed on the switch. • OpenFlow defines a stateless forwarding model for the switch and delegates stateful logic to the controller. As a consequence, realizing SDN-based monitoring presents significant challenges both in terms of scalability and accuracy in mobile networks. Indeed, the network controller needs to directly perform the necessary measurements on each of the numerous (up to tens of thousands) and distant (up to hundreds of kms) network nodes and links with the required precision and granularity (up to microseconds) in order to provide the necessary reliability (up to 99.9999%) to the very distinct services in 5G. To overcome those issues, current SDN-based monitoring approaches needs to be augmented with an automated communication process, namely telemetry, by which measurements and other data are: (i) generated and collected locally at network nodes subject to different service requirements, and (ii) transmitted to the controller for enabling an optimal network management. The goal of this article is therefore to: (i) analyze the mismatch between current OAM tools and SDN solutions in mobile networks, and to (ii) propose a telemetry solution for software-defined mobile networks capable of adapting to the various service requirements expected in 5G. The remainder of the article is structured as follows: Sect. 2 provides an overview of SDN/OpenFlow, OAM and 5G networks for those readers not familiar with these concepts. 1 Afterward, Sect. 3 reports the analysis of the various issues in building an SDN-based monitoring in mobile networks. Next, Sect. 4 presents our solution which is experimentally evaluated in Sect. 5. Finally, Sect. 6 draws the conclusions.
doi:10.1007/s10922-020-09524-1 fatcat:yqgu3xco5jhqveyajelicep6cq