StarlingX — the open source edge computing and IoT cloud platform optimized for low latency and high performance applications — is available in its first release. The project was established in May as a pilot project supported by the OpenStack Foundation (OSF) and builds on code contributed by Wind River and Intel Corporation.
StarlingX delivers services to help fill the gaps in the open source edge cloud ecosystem to meet requirements of edge use cases and scenarios in both carrier and industrial applications.
Additionally, StarlingX leverages components of projects including Ceph, OpenStack and Kubernetes and complements them with services such as configuration and fault management with a focus on key requirements like high availability (HA), quality of service (QoS), performance and low latency.
Use cases reflect the need for high-performance edge and IoT infrastructure
- Low-latency 5G and industrial IoT use cases including industrial automation, cloud radio access network and smart city/buildings (metering and monitoring),
- Multi-access edge computing applications including augmented and virtual reality (AR/VR) and location-based retail,
- High-bandwidth, large-volume applications including mobile high-definition video, content delivery and caching, and surveillance,
- Enterprise-focused small cell services for stadiums and high-density locations,
- Universal customer premise applications (uCPE).
An engaged, growing community
StarlingX release reflects strong growth in the StarlingX community, with 1,329 commits from 84 contributors, including developers representing Wind River, Intel, 99Cloud, China UnionPay, SUSE, Fujitsu and NEC, among others.
Configuration management — The code offers node configuration and inventory management services with auto-discovery and configuration of new nodes, critical for deploying and managing large numbers of remote or hard-to-access sites. The Horizon graphical user interface and a command-line interface manage the inventory of CPUs, GPUs, memory, huge pages and crypto/compression hardware.
Fault management — Users can set, clear and query custom alarms and logs for events on both infrastructure nodes and virtual resources such as VMs and networks. Users can access the Active Alarm List and Active Alarm Counts Banner on the Horizon GUI.
Host management — The software provides lifecycle management functionality to manage host machines via a REST API interface. This tool detects host failures and initiates recovery by providing monitoring and alarming for cluster connectivity, process failures, resource utilization thresholds and hardware faults. The tool also interfaces with the board management controller for out-of-band reset, power-on/off and hardware sensor monitoring and shares host state with other StarlingX components.
Service management — StarlingX provides lifecycle management of services by providing availability through redundancy models like N+M or N across multiple nodes. The service supports the use of multiple messaging paths to avoid split-brain communication failures as well as active or passive monitoring to specify the impact of a service failure with a fully data-driven architecture.
Software management — This service allows users to deploy updates for corrective content and new functionality with a mechanism applicable for all infrastructure stacks from the kernel up to OpenStack services. The module can perform rolling upgrades including parallelization and support for host reboot allowing the moving of workloads off of the node by using live migration.
“When it comes to edge, it is crucial to be able to blend together and manage all the virtual machine and container-based workloads and underlying bare metal environment. This is exactly what you get with StarlingX.”, said Jonathan Bryce, executive director, the OpenStack Foundation.
“Intel is proud of our contributions to StarlingX and work to build a resilient Cloud infrastructure for edge, industrial Internet of Things (IoT) and telecom applications. Today marks a major milestone with the release of the full StarlingX software package. Congratulations to everyone involved for delivering an open, distributed edge solution that supports high-availability and ultra-low-latency.”, said Imad Sousou, corporate vice president, Intel Corp., and general manager, Intel Open Source Technology Center.
“Recognizing the vast need for a virtualized edge solution that can span across emerging 5G and IIoT use cases, Wind River contributed key technologies from our Wind River Titanium Cloud virtualization platform to the StarlingX project. With the first release of the StarlingX project, the community now has a complete, deployable platform that can be used in multiple edge use cases. We are looking forward to even greater participation and accelerated progress from the community to continue to drive new innovation and services deployed at the network edge.”, said Glenn Seiler, vice president of product management and strategy, Software-defined Infrastructure, Wind River.