Skip to main content

A cloud-native vision for SBC

There are several important factors communication service providers (CSPs) need to consider if they are to take advantage of emerging opportunities and achieve their cloud migration and network-evolution goals. In this article, we’ll focus on three key factors.

First, CSPs need to embrace and implement DevOps practices to enable continuous development and deployment of new services and applications. Second, they need to optimize their cloud architectures to ensure they’re making the most efficient use of network resources. Third, they need the agility to leverage new and emerging network technologies like software-defined networking (SDN) and vector packet processing (VPP) to optimize data plane performance.

Meeting these three requirements is at the heart of Nokia’s cloud-native product strategy over the next two years. The Nokia SBC for cloud product provides a prime example of what that strategy looks like in the session border controller domain.

Support for micro-services and DevOps

Traditional network element software is extremely complex, often maintaining customer provisioned and call state data, making it difficult and costly to modify and upgrade. That’s a problem for most CSPs given that the pace of change is eventually going to demand nearly continuous updates to software components.

Refactoring the SBC Virtual Network Function Components (VNFCs) into a collection of smaller, independently configurable micro-services help overcome this challenge. Because micro-services perform basic functions that are stateless, dataless and managed via well-defined APIs, they’re much easier to upgrade and replace in a running system. Micro-services can also be deployed in containers for added portability, further simplifying network evolution and scaling.

With micro-services, customizations can be delivered quickly, easily and independent of the traditional “major releases”. Some examples are:

  • Adapting to emerging security threats and requirements
  • Responding to Lawful Intercept (LI) requests to support changing country specific regulatory requirements or to interwork with deployed 3rd party LI systems
  • Providing Call Detail Records (CDR) changes to align with new OSS/BSS interfaces or matching the CDR formats for legacy SBC replacements.

Cloud network optimization

While the deployment of virtualized network elements consisting of VNFCs in the cloud reduces deployment complexity and CSP OPEX, there has not been a corresponding simplification or optimization of the network. With the migration to a micro-services architecture, new possibilities emerge.

First, a micro-services-based SBC introduces better scalability. Traditional VNFC’s do provide some ability to scale vertically, by the assignment of more resources to the individual VMs, but this scalability is limited by the complexity of the VNFC. Micro-services, being dataless and stateless, can scale horizontally thus providing superior scalability in real time. Micro-services also allow for better optimization. Individual SBC capabilities like firewalls, session management, routing, or transcoding implemented as micro-services can scale independently to more precisely meet specific network needs.

In the longer term, micro-services can also yield better optimization at the network level. Operators may decide to deploy a “decomposed SBC”, centralizing certain functions like transcoding or lawful intercept while distributing other functions like firewalling, pinholing or network adaptation functions closer to the network edge. Managing and engineering some resources like transcoding centrally can be much more efficient than doing so across tens or hundreds of instances in the network.

Enabling new and emerging network technologies

Classic network technologies have tended to be closed, which in the past made it difficult to integrate new functions. For the traditional SBC, this led to implementing required data-plane bearer capabilities in overlay border gateways that were separate from the IP routing layer. This approach, while necessary in the past, runs counter to the real-world needs of today’s cloud networks, which must be able to flexibly configure both control-plane and data-plane capabilities for signaling, call control and bearer functions.

New networking technologies like Software Defined Networks (SDN) will enable a change to this approach. SDN allows operators to create flexible “service chains” of data-plane functions. Nokia envisions that in the future, SBC data-plane bearer border gateway functions can be directly integrated into these SDN service chains, leading to significant optimization of an operator’s network. Additionally, Nokia leverage new development ecosystems like the open-source Vector Packet Processing (VPP) project to guarantee that these new SDN functions will have the necessary line-rate performance characteristics.

From cloud-ready to cloud-native SBC

Nokia has continued to strengthen its software portfolio with industry-leading cloud solutions and in February, announced the third part of our SBC portfolio evolution: Nokia SBC for cloud. The latest release aligns with best practices for cloud architectures as noted by David Snow, GlobalData analyst: “Nokia is taking an early lead in vSBC deployments, supporting media plane processing at scale and including NFV MANO support.”

Nokia continues its commitment to evolving Nokia SBC VNF, including support of the three critical and emerging cloud-native requirements outlined above. The evolution of Nokia’s cloud-ready SBC demonstrates Nokia’s commitment to a cloud-native future and to meet the needs of CSPs as they transform their networks and operations.

For more information on the Nokia Session Border Controller:

Mike Morris

About Mike Morris

Mike has 35 years of experience in the telecom industry. He was chief architect for Alcatel-Lucent’s IMS portfolio, directing IMS evolution to the cloud. His current responsibilities at Nokia include evolving Nokia’s SBC and Enterprise communications offers to a cloud-native environment. Mike received his MS/MA in Computer Science and Mathematics from Michigan State University.

Article tags