The open source StarlingX project is out with its 9.0 release, promising improved performance and capabilities for users of the telco and edge-focused cloud computing platform.

The StarlingX project  is part of the Open Infrastructure Foundation, which is the home of the OpenStack cloud computing platform. In addition to OpenStack, StarlingX integrates Kubernetes and Ceph to help enable edge computing to telco use cases. The new update follows in the same direction of the StarlingX 8.0 update, which came out in Feb. 2023.

StarlingX is already being used in production by major telecom operators around the world. KDDI, T-Systems, Verizon, and Vodafone are among the telecom companies using StarlingX to power their 5G and open RAN (O-RAN) network deployments.

“The main theme for StarlingX 9.0 is strengthening the platform in several areas that are crucial for modern software infrastructure platforms, which include security, robustness, sustainability, and more,” Ildikó Váncsa, Director of Community Open Infrastructure Foundation, told SDxCentral. “As the platform has been in production at several large telecom operators already fulfilling 5G and O-RAN requirements, the community now has the chance to dive deeper into certain areas to further enhance StarlingX.”

For large scale telco deployments having the highest levels of reliability is an operational requirement. Typically the best way to ensure the best reliability is to have lots of redundancy to help account for any potential issues.

Váncsa said that StarlingX 9.0 provides even more built-in redundancy than it has in the past. The update includes features such as redundant system clocks for precision time protocol (PTP) and geo redundancy to increase robustness within geographical regions of a  StarlingX deployment.

The platform was also enhanced with features such as root cause analysis for subcloud deployment and upgrade failures and better configurability for the Forward Error Correction (FEC) feature for all Intel hardware accelerators.

“Resource consumption and sustainability are key for current infrastructure platforms, especially with resource-intensive use cases like AI,” she said.

To help support and improve resource consumption, the StarlingX community added support for Configurable Power Manager, which provides the possibility to set power configuration and regulate the frequency of CPU cores.

O-RAN is a primary use case for StarlingX and in the 9.0 update, the enhanced redundancy capabilities will help to improve operations.

“Since the community has been collaborating with the O-RAN Software Community for quite some time now, they have already established compatibility between the platform and O-RAN requirements,” Váncsa said. “This allowed contributors to focus on configurability and robustness, which makes it even easier to deploy, operate and manage geographically distributed deployments of the platform.”

She noted that the StarlingX platform now also has a Node Feature Discovery (NFD) component. The NFD allows operators to detect available hardware features on each node in a Kubernetes cluster, which improves the configurability of the platform for operators who often rely very heavily on the underlying hardware infrastructure.

Looking forward, Váncsa said that the StarlingX community is now in the phase where tit can focus on fine-tuning the platform and making sure that it delivers features and functionality users need and rely on.

Among such features in the development pipeline are IPv4/IPv6 dual-stack support, which was just discussed during the recent Project Teams Gathering event, and is planned for StarlingX 10.0.