Network as a Service is a cloud computing model in which managed cloud vendors provide virtualized network equipment and networking functionality as a subscription service. A concept similar to the better-known software/platform as a service model, Network as a Service lets enterprises operate networks without actually owning and maintaining the hardware.
Network as a Service vendors provide core network elements such as routers, firewalls, MPLS connections and application delivery control in a virtualized form (replacing hardware with software). These are delivered as pay-as-you-go services, effectively turning networking into an operational spend rather than a capital investment for enterprises.
Network as a Service use cases include a multitude of connectivity solutions such as virtual private networks, wide area networks, data center connectivity and bandwidth on demand.
How is Network as a Service deployed?
Network as a Service deployment is in the early stages. The technology—and the uses we find for it—are rapidly evolving.
Layer 4-7 functionality (transport, session, presentation, and application) is already widely available as virtualized services. These cloud computing models are deployed extensively and have proved their usefulness beyond any doubt.
Layer 1-3 functionality (physical, data link, and network) is a little more complicated because physical hardware is often required to handle data flowing between endpoints and the datacenter. To address this, enterprises can choose to deploy Network as a Service alongside legacy infrastructure and we often see such mixed deployments in the field.
Currently, enterprises with multiple branch locations and remote workers (work from home or in-the-field users) use Network as a Service to enable WiFi control, SD-WAN, perimeter and endpoint security, application control, and other overlay networking services through the cloud. Implementing Network as a Service lets corporate employees access internal resources through a web browser over an internet connection. The Network as a Service provider secures their activity, protects their data, and routes the web traffic wherever it needs to go.
5G and Industrial IoT – with their proliferating smart device connections, distributed workloads, and ultra-low latency demands – will play a vital part in accelerating Network as a Service deployments. 5G’s predictable single-digit millisecond-latency, capacity to support 1 million devices per square kilometer, and network slicing capability make it an ideal platform to deliver Network as a Service.
Moving forward, we will see more hybrid Network as a Service offerings evolve. These are expected to take the form of virtualized options for multi-cloud control, third-party access, network automation, cloud intelligence, and management of on-premises hardware.
The history of Network as a Service
To capture the full potential of Network as a Service, it’s important to step back and see why (and how) this cloud service evolved.
Traditionally, enterprises have distrusted the public internet as a means of operating their own businesses. Instead, they put their faith in private internal networks (also known as corporate networks or intranets) that connected different parts of the business and allowed employees in different locations and functions to access internal data and communicate with each other. So enterprises set up their own wide area networks, provisioned each branch with perimeter protection (firewalls, DDoS mitigation, etc.) and connected the different branches using rented/dedicated connections (such as MPLS).
The company’s private information and the systems (tools, processes, workflows, etc) needed to access it resided in the closed private network and employees could access the data and systems by logging into the corporate network using secure access protocols (such as VPN). Any branch employee who needed to access the internet would first traverse through the entire corporate network—from branch (nodes) to headquarters (datacenter)—and then onto the public networks.
The most frustrating part of this exercise was the lack of speed. First, having to navigate the entire network and tech stack added unnecessary data roundtrips and slowed down the traffic. Second, any bottlenecks anywhere in the network could slow down the entire network.
The corporate network set-up—while secure—quickly proved to be inconvenient and inefficient for everyone concerned:
- The enterprise had to make massive investments in hardware (the network infrastructure), software (to control the network), and hiring in-house IT resources (to ensure it all runs smoothly).
- Employees become less productive and more stressed because of the time and effort it took to access the information and tools they needed to do their jobs,
- IT teams needed to be on top of latest releases and versions of the software. They also needed to police the system to ensure employees followed security protocols and installed the updates and patches needed to keep the network running smoothly. Time they could have used to improve the system and innovate was lost to routine maintenance tasks.
Even with all these challenges, enterprises learned to live with the traditional corporate networks. Until cloud computing came of age.
Once enterprises became comfortable with the cloud services model, it was only a matter of time before Network as a Service was conceived. With the technology advancements of the past decade, now the entire network can be offered as a utility, instead of just network pieces (such as software or data infrastructure).
Network as a Service vs. Private 5G Networks
Private 5G networks are physical or virtual 5G networks that are exclusively deployed for the dedicated use of governments or enterprises. It can be an on-premises physical network, a virtual private network, or a private slice of a 5G network. Private 5G networks are deployed to guarantee network availability for mission-critical functions (such as public safety or robotic surgery) or business-critical use cases (think fleet management, smart factories etc.).
Experts opine that Network as a Service and Private networks will coexist, likely as a hybrid deployment model. In such deployments, sensitive operations and data can be stored in the private network while the underlying core network services (such as BSS/OSS) can be a Network as a Service component. The combination delivers the most critical modern network capabilities—predictable millisecond latency, high availability, robust security, and operational resilience.
Advantages of Network as a Service
Network as a Service offers all the advantages of other cloud-native services, including:
- The flexibility to rapidly introduce and modify network services
- Rapid provisioning that aids the ability to scale as quickly as business needs change
- Seamless installation of the most current and secure software updates
- The freedom to access the network from anywhere
For enterprises, Network as a Service is an opportunity to free up in-house resources to focus on running and scaling business operations without having to worry about the underlying networking functionality. It also immensely improves the employee experience and makes their teams more productive and collaborative.
For telcos, Network as a Service is a chance to monetize their 5G investments and substantially increase their return on investment by extracting more value from the network. It also represents a step into the future of networking. The digital transformation of networks – supported by agile BSS/OSS systems and a data platform that meets 5G’s scale, latency and transaction consistency levels—is the foundation on which emerging technologies will deliver value.
VoltDB – the modern data platform for modern networks
Network virtualization heralds a new age of real-time, event- and data-driven intelligence that requires handling complex decisions leading to automated operational optimization at the speed, scale, and precision that 5G demands. VoltDB is a built-for-5G fast data platform that combines the velocity of streaming data with the state-based consistency of an operational in-memory database, and the decisioning intelligence of advanced analytics.