Distributed cloud and edge cloud applications will fuel the next generation of open source innovation
Paul Miller, VP for industry solutions at tech specialists Wind River, looks at how open source is evolving the way telcos operate
Open source innovation
5G and its emerging applications are creating the need for a geographically distributed virtualised infrastructure onto which these applications can be hosted. In Open Source, we are seeing and supporting significant innovation in the OpenStack foundation’s StarlingX.io project, where work is ongoing to create a cloud native, but geo-distributed infrastructure that meets these requirements. This is a challenge that legacy open source projects, by themselves, cannot meet the need. Thus, the next generation of open source cloud technology is focused on distributed cloud, and edge cloud type applications.
At the same time, open source state of the art is also seeing a transition from virtual machines to containers, as a much more performant, scalable, and resource efficient solution for virtualisation. There will be a need to take the best of breed virtualisation technology, like open source based Kubernetes, and use that as a foundation to create an end-to-end cloud solution that enables and embraces the challenges of a geo-distributed network architecture. This creates an intersection between the demand for next generation distributed cloud, and the emergence of cloud native approaches for virtualisation.
As a result, interesting innovations will continue to emerge related to increasing the adoption and usability of containers. This will occur in the areas of securing container workloads and also in autonomous operation and management of workloads at far edge sites. This can mean use cases such as Multi-Access Edge (MEC) applications pushed out to remote far-edge locations in MEA. The technology advances in container security, container management and overall operations from centralised locations out to the far edge. Farther out, the adoption of machine learning and AI into these same use cases will be the next leading trend.
The role of edge computing
Telecom infrastructure is evolving from vertically integrated, monolithic solutions to disaggregated technology based on virtualisation and the cloud. The multifaceted needs of a 5G network create new requirements, both technological and operational, that existing cloud infrastructure cannot meet. Due to 5G’s requirement to distribute compute nodes to thousands of different sites to support the explosion of new connected devices, there is a fundamental mismatch to legacy virtualisation approaches, especially around operational concerns. vRAN Technology based on cloud native, geo-distributed cloud solutions is ideal for solving the modern problems, both technological and operational, introduced by 5G.
Using these new approaches, telecom service providers can offer new capabilities to industries that will rely on edge computing such as automotive, industrial, and gaming. Telecom service providers can support on-premise solutions as-a-service for industries that require the attributes of 5G, as well. The possible answers to these challenges have telcos eager to build a network to support the promise of 5G that ensures:
- Greater performance
- Ultra-low latency
- More comprehensive security
- Uncompromised reliability
- Intelligence (AI, analytics)
- Open source-based technology to maximise flexibility
- Operationally friendly ways to deploy/manage complex, large, distributed cloud networks as they want to rapidly deploy new services
- Improve OPEX and CAPEX (through virtualisation on COTS vs. integrated hardware)
Thus, edge computing is the foundation of the next generation service provider network, not only hosting the 5G infrastructure, but also hosting applications for next-generation use cases such as vehicle to vehicle communications for accident avoidance, autonomous self-driving cars, automated industry, remote telemedicine, autonomous drone control and more.
More mission critical systems across markets are becoming connected, and failure is not an option. New intelligent systems will have diverse needs around the cloud, virtualisation, performance and latency, security, reliability, and intelligence. In this new paradigm, it will be important to seek experts, like Wind River, who have built solutions from the ground up specifically to meet the unique needs of telecom environments.
Use cases in areas such as industrial automation/IIoT, avionics/commercial drones, autonomous vehicles, and medical systems are interesting as they involve high stakes and demand more intensive requirements, specifically in regard to low latency, smaller footprint, and reliability. Latency will play an ever-increasing role in the ability for 5G deployments to function, scale, and cost reduce. We’ll see the emergence of applications that are fully software based that are dependent on ultra-low latency (URLLC) access from the edge to the core to enable the application to function properly. Drawing from its experience in mission critical industries (telecom, aerospace and defense, industrial, automotive, medical), Wind River understands and has the expertise to address the technology requirements that 5G demands.
Opportunity in the Middle East and Africa
As Wind River is very focused on open source based, virtual edge computing platforms specifically focused on 5G use cases, and the MENA region by 2025 is anticipated to have more than 50 million 5G connections with networks covering 30 percent of the population, we see an incredible opportunity to bring our technology to this market and help service providers solve the serious challenges associated with 5G network planning and deployment.
We anticipate the GCC states to be among the first to launch commercial 5G networks in the region, including efforts by Etisalat and Ooredoo with both companies working on 5G trials moving towards commercial launches. We keep a close eye as well on regulatory paradigms associated with spectrum allocation for 5G, as any issues here can gate early deployment. We see the UAE within the GCC in MENA as likely having the highest volume deployments of 5G.
Thus, we see the MENA region tracking the massive global movement and investment in 5G, and in a similar manner to other regions, find our technology uniquely adept at solving problems associated with 5G deployment.
What predictions do you have for the industry over the next 12-18 months?
While early deployments largely focused on early adoption and trials of the 5G RF infrastructure have been built with appliance based, legacy hardware, we see the future of 5G deployment at massive scale to be founded on a virtualised, disaggregated approach. Within the next 12-18 months we anticipate the maturation of 5G technology combined with a global shift to employing containerised virtualisation as the platform of choice to host this infrastructure.
As 5G matures, we see MEC technology following, leveraging the distributed cloud technology that has been deployed for 5G infrastructure, now being scaled to host third party applications as is associated with MEC use cases. These includes ultra-low latency applications that require hosting at the extreme edge of the network, enabling commercialisation of autonomous systems and vehicles, automated factories and remote real time telemedicine to name a few.
Thus, the evolution to 5G and advanced applications, enabled by this new distributed cloud infrastructure creates the next-generation service provider network. One fully virtualised, orchestrated and automated, with revolutionary flexibility and ease of use, providing rapid time to market for new revenue streams.