Why containers, white-box devices, and SD-WAN are perfect for cloudifying the network edge

Posted by Marbenz Antonio on May 16, 2022

White boxes in the enterprise: Why it's not crazy | Network World

Containers, software-defined wide-area networking, cloud edge communications, and white-box devices

Compared to dedicated bare-metal solutions, containerized Software-Defined Wide Area Networking (SD-WAN) on white-box x86 platforms offers a higher degree of flexibility and similar performance — especially when the goal is to deliver dynamic edge compute solutions with more outstanding security capabilities.

Container technology allows software workloads to be deployed anywhere, allowing telecom service providers and others to construct flexible SD-WAN networks that improve the security profiles of multi-cloud and edge cloud communications. Corporate applications in numerous public and private clouds might be connected with appropriate edge compute devices or uCPE at branch offices, weather-proof nodes connected to IoT devices, or 4G/5G edge devices in fleet cars, trucks, and heavy equipment, or boats, for example.

Edge computing devices are easier to acquire and manage using white-box technology, which reduces outlays for spare parts and recurrent maintenance expenses, while container technologies enable deploying or upgrading software in edge devices simpler and easier.

A multi-vendor setup for performance testing containerized SD-WAN

Intel, Red Hat, Turnium, and TietoEvry partnered on lab-based testing in the fall of 2021 and early 2022 to demonstrate the benefits of containerized SD-WAN utilizing Turnium’s SD-WAN Cloud Native Network Function (CNF), which is certified on Red Hat OpenShift and runs on Intel-powered servers. To implement, maintain, and test this software stack, TietoEvry offered a third-party testing environment, traffic simulators, and technical employees.

The purpose of the test was to see how dedicated, fixed-function deployments compared to software-defined, containerized deployments performed.

The performance of two edge types of technologies was compared in the research:

  • A cloud-native containerized deployment of Turnium’s SD-WAN as a containerized application deployed, controlled, and managed by OpenShift on an x86-based Universal CPE (uCPE).
  • Turnium’s SD-WAN software image was immediately deployed on top of the uCPE operating system in a bare-metal deployment.

These edge nodes, also known as uCPEs, interacted with core nodes using Turnium Aggregator software, which was installed on:

  • A public overlay network connects the aggregators and uCPE, simulating a public/private cloud offshore from the edge nodes.
  • A simulated local aggregator is located on the same private network as the edge nodes or uCPE.

Test results and concluding remarks

The performance testing revealed that the containerized SD-WAN met the needed performance while also providing improved scalability, flexibility, simplicity of deployment, and automation for large-scale edge computing solutions.

It was simple to add new nodes to the network automatically throughout the test. Automated network programming and features, such as wireline and wireless bonding and failover, made it possible to swiftly set up a resilient network. Deploying OpenShift demonstrated that controlling numerous edge devices at scale was a trivial process, and separating the software from the hardware using commercial off-the-shelf (COTS) hardware made it easy to size the device for core and edge nodes.

In terms of performance, tests revealed that OpenShift had a little influence on system load, adding about 2% more CPU demand as compared to running the SD-WAN implementation on bare metal. This low marginal CPU effect allows companies and corporations to profit from edge compute and public or private 5G deployments by allowing them to build and run more containerized applications within the uCPE.

By combining computing at the edge with technologies like Turnium’s containerized SD-WAN platform and Red Hat OpenShift, it’s easy to quickly install workloads or apps to take advantage of the low cost and high processing power of x86 COTS equipment.


Here at CourseMonster, we know how hard it may be to find the right time and funds for training. We provide effective training programs that enable you to select the training option that best meets the demands of your company.

For more information, please get in touch with one of our course advisers today or contact us at training@coursemonster.com

Verified by MonsterInsights