5G triggers real hype among providers of digital services. New services are being brought to market faster and faster, and networks are being optimized in terms of their performance to be competitive in the emerging 5G ecosystem. Extremely important in this context: Network Functions Virtualization.
The virtualization of network functions ( NFV ) and software-defined networks ( SDN ) play a key role in the design of their offers and potential application scenarios) an important role. But even if extensive NFV implementations have been carried out, these usually only apply to a small part of the network operator’s data traffic. The NFV is undoubtedly still the most pragmatic and fastest way to introduce new services for companies and entire industries. Additional capacity for IoT applications can also be realized using NFV. Service providers are therefore well advised to continue to deal with NFV, as it increases connection speeds at the age of 5G.
It is widely believed that most companies are overwhelmed by the parallel orchestration of NFV and a traditional resource management network. Therefore, many operators revise their calculation of the return-on-investment of an NFV deployment and consequently reduce the scaling.
To date, several leading service providers have not yet fully implemented NFV and only route relatively small volumes of their data traffic via their virtual infrastructures. Nevertheless, they want to introduce new 5G services as quickly as possible as the technology becomes more and more important. At the same time, they are unsure whether NFV can help them achieve this goal. They underestimate that virtualization is the most efficient strategy and NFV brings a plus in scalability and agility that can benefit growth.
Accordingly, the Federal Ministry of Education and Research in its white paper “5G Characterization” states that software-defined networking (SDN) and network functions virtualization (NFV) are key features of 5G, through which the entire network has a programmable, flexible and universal infrastructure receives.
In multi-cloud environments, applications and workloads are distributed across different cloud environments and often switch between environments at short notice. Service assurance systems must, therefore, provide a correlated view of the performance of the applications and the networks used across the WAN ( SD-WAN and IP / MPLS ).
An almost real-time, highly granular view of application and network performance makes it possible to detect deteriorations in performance and take corrective action before users are affected. For this, the assurance system must work together with the SD-WAN controller and the orchestrator in a closed control loop to receive real-time information about the service chains within the uCPE (Universal Customer Premises Equipment) of the selected dynamic paths and to provide the controller and the orchestrator with performance information To make available. This enables a detailed analysis of the impact of network performance on application performance to proactively identify, isolate, and prevent service degradation.
In The Clouds
Cloud providers are cooperating with Carrier Service Providers (CSPs) to reduce the complexity at the digital edge – i.e. at the edge of the networks – that results from the switch to 5G. The so-called EdgeComputing bypasses the bottleneck of ever larger in the transmission data amounts of central processing units arises, eliminating a major cause of propagation delays. This is a basic requirement, especially for applications that rely on large amounts of data in real-time, such as connected cars.
Cloud providers also package and monetize public architecture to accommodate systems that require less latency and massive throughput. NFV architectures offer the necessary network flexibility for new service models and fast scaling. However, these advantages can only be realized if the NFV-based solutions are integrated via toolchain technologies for automation and orchestration.
The Solution Virtual And Intelligent
What service providers need is intelligent transparency for a distributed architecture. Software solutions are available for this purpose, which collect relevant data using virtualized probes consistently and across domains for all network layers and transmit them to analysis tools. It is recommended to use a network analysis platform for the central aggregation and correlation of the performance data of the network and applications, which has freely definable reporting functions and thus guarantees reporting. Advanced machine learning-based analysis functions also provide forecasts of service quality so that deteriorations can be recognized and contained in a good time.
Virtual probes with intelligent data technology and the corresponding tools record and analyze all important data points to provide usable knowledge in practice. “Intelligent” is of crucial importance in this case and stands for not transferring all information into the analysis flow without prioritizing, but ensuring that the data is as relevant as possible to the functionality. Because the processing of large amounts of irrelevant information is not only costly but also time-consuming, which precludes an immediate reaction to incidents.
Rapidly responding and precisely working data tools, which consolidate information according to the specific operational framework, form the basis for being able to react to service failures almost in real-time. 5G has other advantages as it increases the ability of a service provider to understand what is going on in the network and thereby gain insight into the performance characteristics of a particular service. For example, this could be an SD-WAN service for companies in the early phase of 5G provisioning or, shortly, a high-revenue consumer service such as UHD gaming.
To achieve these results, service providers must of course increasingly use virtualized infrastructure for software-defined solutions. 5G in particular needs virtualization to ensure not only performance but also the economy. And virtualization needs 5G because it offers the opportunity to demonstrate the advantages of a fully virtualized operating environment for the first time.