We all know what cloud networks are. It is IT infrastructure in which some or all of an organisation’s network is hosted in a public or private cloud platform, that is either managed in-house or by a service provider, and available on demand.
Cloud metro networks are an evolution of this, as a convergence of cloud computing, edge computing, 5G and metro networks.
In his own words Vikram Nair, president of EMEA business at Tech Mahindra, characterises similarly saying, “cloud metro networks refer to the latest metropolitan area networks (MANs) designed to support the low latency and high bandwidth needs of cloud-based applications and services.”
“Cloud metro networks enable service providers to build scalable, secure, and high-performance networks that help them meet their sustainability objectives by reducing energy consumption and CO2 emissions.”
Described as an “experience-first network” it is where connectivity, distributed cloud, and customer experience converge. And while cloud metro networks simplify deployment for service providers, it also “enhances efficiency, agility, and scalability, surpassing the capabilities of traditional metro network infrastructure”, he says.
Expanding on this, Julius Francis, senior director of product marketing & strategy at Juniper Networks, says that Metro Networks are underpinned by three key sustainability elements: operations, systems, and architecture.
“Operations are built with automation that is AI-enabled and cloud-native by design. Systems deliver port density and flexibility speeds to 400G while reducing carbon footprint and e-waste. Cloud Metro’s architecture has a scalable IP services fabric with embedded active service assurance and zero trust security,” he says.
Metro traffic bandwidth is predicted to grow by more than 500% between 2021 and 2027, with it comes not only increased demand but also new consumer and enterprise use cases.
According to Jürgen Hatheier, CTO of EMEA & APAC at Ciena, this is driven by the fact that cloud metro networks focus on high-speed, low-latency connectivity to ensure optimal performance for cloud applications.
These often require real-time or near-real-time interactions enabled by connecting to edge data centres located closer to the network edge, where content is created and consumed.
As a result, “cloud metro networks ensure smoother and more responsive streaming, online gaming, and immersive AI/VR/AR experiences for consumers,” he says.
“On the enterprise side, these networks support edge computing for smart cities, as well as in industries such as manufacturing and retail. The next stage of evolution for these networks is just around the corner with AI taking the application space by storm.”
Robert Gazda, senior director of the wireless & networking lab at InterDigital, says thanks to the unique features of metro cloud networks, such as programmability, virtualisation, autonomous / dynamic management, these new uses can be explored.
“The programmability brings new capabilities where the network is more aware of application packets and operates efficiently in processing/forwarding/steering traffic according to application requirements,” explains Gazda.
“These capabilities provide a better quality of user experience and will enable immersive applications, such as interactive mixed reality and the metaverse, which require ultra lower latency with high levels of processing (in-network and at the edge).
In addition, cloud metro networks can provide virtualised links, virtualised topologies, and virtualised metro networks enabling sliced metro networks and networking sharing.
“For example, a smart factory enterprise that deploys a local 5G-based standalone non-public network can lease a metro network slice for connecting the standalone non-public network to the enterprise’s cloud servers. Multiple MNOs can also share a physical cloud metro network to reduce investment and operational costs,” he says.
One of the more highly touted advantages of cloud metro networks is greater sustainability, specifically greater energy-efficiency and innovative power designs.
Chris Sharp, CTO at Digital Realty, expands on this saying, “sustainability has become a top agenda item for consumers and companies alike and cloud metro networks are at the forefront of driving greater energy efficiency in digital infrastructure.
“By integrating software-defined networking (SDN) and network function virtualisation (NFV), as cities continue to grow these networks can dynamically adjust to traffic demands, thus optimising energy usage.”
He refers to the recent news of BT modernising much of its global network fabric, which is based on a network-as-a-service (NaaS) technical and commercial model and connects the multiple clouds businesses use for their applications and data. This new NaaS, called Global Fabric, has reduced energy usage and carbon emissions by approximately 80%.
“These networks virtualise individual network paths through a larger overall network fabric, the amount of networking hardware required per customer is reduced compared to traditional per-customer network architectures,” added Sharp.
On more of the hardware side of things, Mike Hoy, technology director at Pulsant, commented: “many readers will remember the green cabinets full of patch cables, connecting pairs of wires to houses and the exchange. Now, with cloud metro networks add active infrastructure at the edge, making it more energy efficient.
“Instead of large network equipment at big data centres, old tech is being swapped for new, energy-efficient gear from DC to edge, making the whole setup more sustainable.”
AI-enabled, cloud-delivered automation also plays a key part in cloud metro networks, particularly with its experience-first networking approach.
“In cloud metro networks, AI-enabled automation is a better, faster and simpler way to achieve experience-first networking,” says Francis.
“This intuitive user experience, combined with state-of-the-art intent-based networking, simplifies and accelerates service delivery, reducing time-to-service introduction from months to minutes and improving productivity and time to revenue. “
In addition, Sharp says, “AI algorithms can predict traffic patterns and adapt the network resources accordingly, ensuring optimal performance without human intervention, and thus, reducing downtime and operational costs.”
It can also help providers simplify the ability to design and deploy solutions for enterprises by suggesting providers that might provide them value or automating troubleshooting to minimise downtime.
Overall, it seems that the software-based nature of cloud metro networks is its biggest differentiator, meaning a return to device driven networks is increasingly unlikely, and central to the evolution of the network.
“Cloud metro networks are often more software-driven because these networks are designed to dynamically scale to accommodate the growing demand for ever-changing cloud services,” explains Hatheier.
“Traditionally, applications were driven by the performance (and limitations) of the underlying network infrastructure. The inverse is now taking place where the network is evolving based on the requirements of new use cases and applications.”
This virtualised approach also feeds into interoperability of the cloud metro network, making it more open, agile and responsive.
“They’re not tied down to specific devices and are accessible from any proprietary system. It’s all about being flexible and open,” says Hoy.
“Cloud metro networks will decouple software from particular hardware, freeing you from ‘lock-in’ to a particular vendor’s products; different pieces from other organisations can cohesively integrate as a sort of ‘network without borders’, promoting innovation and healthy competition.
With a 5–10-year outlook, Nair expects to see more businesses implement cloud metro networks where 5G, edge cloud hosting, connectivity and service experience converge.
“A new ITU standard now requires operators to reduce greenhouse emissions by 45% by 2030 and if businesses want to enable sustainable business growth, they’ll need to apply cloud principles to their metro networks,” he says.
Sharp predicts that cloud metro networks will integrate with AI and machine learning technologies and that “the networks will evolve to support emerging technologies such as 6G, quantum networking, and more pervasive edge computing architectures”.
Hoy says that “overlay and software networks will take the reins as opposed to traditional telcos”, while MNOs adopt private access point networks “for dedicated, secure network spaces” and “devices will be connected wirelessly or otherwise across virtual MNOs, consolidating mobile and fixed line services – all as a single network, and all driven by software”.
Gazda points to the move from 5G to 6G as another contributor to the future of cloud metro networks likely driven by new use cases and network scenarios which may require new or evolved capabilities for cloud metro alongside access networks and devices.
One such example is AI-generated content (AIGC) which is quickly growing as a new type of network traffic, which he says not only requires higher communication, but also the integration of communication and computation.
“AIGC may be generated anywhere in the network, including the device and edge, but especially in the cloud metro. As a result, future cloud metro networks must support flexible and efficient communication and computation integration including in-network intelligence,” he explains.
As for 6G, we anticipated that future 6G access networks will be more distributed and integrated (e.g., space-ground integrated network) compared to 5G.
“To better support such distributed and integrated 6G access networks, Future cloud metro needs the capabilities of facilitating distributed communication and computing within access and metro networks. Future cloud metro also should be evolved to support space access networks,” concludes Gazda.
Lastly, Hatheier says that security will remain a top priority for cloud metro networks.
“Over the coming decade, we will see the implementation of advanced security measures, including improved encryption, threat detection, and response mechanisms to combat the evolving landscape of cyber threats.”
