There are many promising new applications being designed to take full advantage of cloud capabilities. However, many of these applications present requirements that today’s networks and operational practices struggle to support.
To seize these opportunities, communication service providers, enterprises, and other organizations must rethink how they design their networks, develop software, run operations, and manage vendor relationships. A shift in architecture, processes, and collaboration models is essential to deliver the agility and scale modern services demand.
Consider the Internet of Things (IoT): while often discussed in terms of sheer device counts—potentially hundreds of thousands to millions of devices per square kilometre—these deployments produce a very different traffic pattern than traditional consumer services. Many IoT applications generate a disproportionate amount of control signalling relative to user data, which contrasts sharply with services such as video streaming where the bulk of traffic is payload data.
Although recent 3GPP enhancements to 4G and 5G standards aim to address some IoT challenges, the underlying network architecture must become far more flexible and scalable to handle scenarios where millions of devices might simultaneously come online in response to local events. This requires mechanisms for rapid, large-scale resource allocation and control-plane scalability.
Beyond IoT, growing investments in automation—ranging from autonomous and connected vehicles to industrial automation, logistics, and smart infrastructure—demand deterministic, low-latency communications. Many machine-critical processes will require end-to-end latencies on the order of 10 ms, while tactile, haptic, and other ultra-responsive use cases may need latencies approaching 1 ms.
Meeting these stringent latency and reliability targets requires distributing compute and intelligence closer to where decisions and controls happen. In practice, this means deploying cloud-native processing at the edge—so-called edge clouds—that bring compute, storage, and specialized services much nearer to devices and sensors.
Simultaneously, immersive experiences like virtual reality (VR), augmented reality (AR), mixed reality (MR), and collaborative multi-user 3D applications demand both very low latency and ultra-high bandwidth. At the same time, demand for on-demand multimedia and video continues to grow, further increasing stress on transport and access networks.
The Specifics of a Cloud-Native Architecture
To build the flexibility and dynamic resource allocation these use cases require, networks must adopt cloud principles. Key concepts include Network Functions Virtualization (NFV) and Software Defined Networking (SDN), applied across a cloud-based infrastructure. These approaches, first proven in data centres, are now being extended into the broader network.
In practice, networks will operate for some time as hybrids of physical and virtual functions, but the trend toward virtualization will persist. NFV enables rapid reallocation of compute and networking resources to new functions, which will be essential for future networks that must support highly variable and unpredictable demands.
When NFV and SDN are combined with cloud-native operations, the result can be highly scalable and flexible networks—but only if the network functions themselves are designed to be cloud-native. This means disaggregating monolithic functions into smaller, composable components; designing state handling to be as stateless as possible; and centralizing or federating data storage so multiple components share common repositories.
Cloud-native network functions enable flexible deployment models across core, regional, and edge locations. Functions can be dynamically chained and interconnected to create tailored services for individual consumers, enterprises, or dense IoT deployments in real time. Crucially, these services should be able to run economically on a common, adaptive infrastructure that dynamically matches capacity to demand.
As different classes of users and applications place diverse demands on the network, business models will evolve as market participants rapidly adopt innovative services. This shift will require organizations to automate more of their operations, adopt intelligent resource management, and embrace agile development and DevOps practices. Close collaboration between operators and vendors will be required to accelerate delivery and maintain operational stability.
Collaboration is Key
In the emerging digital landscape, agility, contextualization, and personalization are vital service differentiators. To deliver these qualities, organizations must pivot from a network-centric to a customer-centric mindset, reorganizing priorities and investments around customer experience and rapid service delivery.
Operational cost optimization will be critical as capital and operational investments shift toward customer-facing innovations and faster time-to-market. Automating the service lifecycle—through continuous monitoring, analytics, and closed-loop assurance—will reduce operational expense and create near-continuous feedback loops for developers, making them more “operations-aware” and better able to produce resilient, deployable code.
As these practices mature, a more unified environment spanning network and IT operations will emerge. IT will no longer simply hand over applications for operations to run; instead, development and operations teams will collaborate closely, sharing operational insights during the development cycle so applications and platform components are built with network constraints and customer expectations in mind.
Reliability in cloud-native networks depends on detecting and remediating defects early across many interconnected, service-chained components. Software will be updated more frequently, so organizations must adopt testing, observability, and automation strategies to ensure updates do not degrade performance or availability.
Vendors also need to be active participants in agile development and platform evolution, since many will build the applications and runtime platforms operators depend on. This requires a deeper, more cooperative relationship between operators, enterprises, and vendors than has traditionally existed.
Transitioning to a cloud-native posture is both a natural evolutionary step and a transformative change for many organizations. It means embracing software-driven operations that are more responsive to customer experience, faster to innovate, and more flexible in how resources are allocated and consumed.
The fusion of the cloud and the network will form a foundational platform for economies and societies in the coming decades. To fulfill that role, organizations must become more flexible, more automated, and better able to collaborate across development, operations, and vendor ecosystems to meet a rapidly shifting set of demands.
What are your thoughts on cloud-native architectures? Let us know in the comments.