The automation journey - step 3: distributed cloud & network slicing

The third step of the journey

Ericsson has been working jointly with customers on a journey that could be presented in three steps:

• Step 1: Scale virtualization
• Step 2: Data Center transformation
• Step 3: Distributed cloud and network slicing

The main challenges for the final step are related to the complexity of a hybrid environment and the need for agility to capitalize on new services introduced with private LTE and 5G. A hybrid environment brings the inefficiency of working in silos (both technical and organizational) and higher operational costs. New service introductions demand faster time to market so that monetization can begin as quickly as possible. However, accelerating monetization has its challenges. Automation can help overcome those challenges by simplifying operations, so the time to market is reduced.

The complexity of the network also relates to the fact that most new applications will be built on cloud native. Micro-services architecture enables speed on the implementation of new functionalities and that is the main reason why new applications make use of this new design. On top of that, new services that require extreme low latency or extreme high bandwidth must be available at the edge to reduce latency and avoid network congestion. Edge sites, as well as public cloud, extend the cloud boundaries, but add complexity to the network. That's why it’s important to consider an orchestration solution that can handle physical, virtual, and Cloud native applications located across a distributed cloud.

The benefits of driving automation in distributed cloud and network slicing

According to a recent 5G business potential study by Ericsson, service providers can expect between 30 percent and 40 percent revenue growth, with new services coming from 5G and IoT. How? Private LTE and 5G will eventually be able to provide new and extremely low latency service offerings, such as remote mining and factory robotics. Network slicing will help service providers to introduce and test those new use cases in a controlled environment. Network slicing automation helps to quickly create and manage hundreds or even thousands of network slices, reducing complexity. Each slice can contain different requirements depending on the type of service (enhanced mobile broadband, massive machine-type communication or critical machine-type communication). Alignment with open source initiatives, such as ONAP, allows the use of templates on activities such as design, creation and deployment of network slices in an automated way.

The introduction of network slicing automation addresses the complexity while continuous delivery and deployment (CD&D) addresses agility. With continuous delivery and deployment, it's possible to offer service providers automated software distribution, automated software deployment, automated acceptance tests and automated collections and analysis of data for feedback to the software vendor. Those automated activities drastically reduce time to market with low risk and higher quality. It's at this point that operators can realize the full potential of these new services.

The use of machine learning and artificial intelligence in orchestration can provide a much more efficient way to do service orchestration. Machine learning uses statistics techniques to improve tasks with experience. Artificial intelligence uses logics, rules and machine learning to simulate human intelligence. Those techniques help to address the requirements of low latency and high reliability applications in a distributed cloud, such as emergency communication and connected cars. Automated service orchestration, with use of machine learning and artificial intelligence, take us closer to a zero-touch network.

Why Ericsson is the best partner for automation and orchestration

Ericsson Dynamic Orchestration is the solution that drives the automation of network slice creation, configuration, and deployment. It implements end-to-end service and transport orchestration that spans multiple domains in a multi-vendor environment. With a single orchestrator for radio, transport and core, the solution provides management of physical, virtual and cloud native applications.

Ericsson Dynamic Orchestration is fully ETSI compliant and aligned with ONAP. It uses machine learning and artificial intelligence in service assurance and analytics for the best service optimization and customer experience.

Watch the video The automation journey – Moving towards zero touch networks featuring Ignacio Más, Head of Strategy in Solution Area OSS, to learn more about the third step of the automation journey.

Read the first blog post in this series, 5G orchestration: The automation journey.

Read Arvind Duray's blog post about Network slicing orchestration.