Let's explain network orchestration, its relation to network automation, and why it matters to your company’s management.
What is network orchestration?
Network orchestration is network automation across multiple types of infrastructure devices, network domains, and even multi-vendor systems in a network. More specifically, orchestration can be policy-based or event-driven automation through programmatic interfaces like RESTful APIs enabled by third party software or within a single integrated end to end network infrastructure solution.
Speaking of network automation, there’s some confusion and ambiguity around the differences between network automation and network orchestration.
What is the difference between network automation and network orchestration?
In simple terms, network orchestration is a more advanced level of network automation. You can think of network automation handling low-level single “if this then that” tasks, whereas network orchestration manages high-level sequences of inter-dependent tasks across multiple systems.
Network automation is often somewhat linear. Since orchestration is more network-aware, it executes workflows based on device states and configurations. Programming is done through command lines or third-party script-driven frameworks. By contrast, orchestration is done through programmatic RESTful interfaces like API calls.
Now that we’ve differentiated the difference between the two, let’s dive a little deeper into the different types of orchestration.
Types of Network Orchestration
There are three main types of orchestration, these include Policy-Based Automation, Software-Defined Networking, and Intent-Based Networking Systems. We’ll explain each and their differences.
Policy-Based Automation (PBA)
Policy-based automation abstracts individual device parameters and instead controls device “policies.” Policies are applied per specific device types or roles. This tends to be the simplest type of orchestration that can pave the way to “software-defined” orchestration.
Through the creation of policies, large organizations can easily bundle devices and services into logical groups that represent key functions of their business. This contextualizes the prioritization of resources at the business level, while still allowing IT departments to understand the underlying architecture that supports those functions.
Policies can normally be copied between devices or supporting systems, and used to create a template that can be implemented at new facilities if the schema remains unchanged. You can often find policy-based automation in GUI dashboards since it’s generally considered one of the easiest ways to implement changes across devices.
Software-Defined Networking (SDN)
Software-defined networking adds a software-based controller or APIs. These controllers allow programmatic management of functions, like provisioning, monitoring, and configuration. This tends to be more sophisticated than policy-based automation but less than intent-based networking systems.
Unlike policy-based automation, software-defined networking is directly programmable and separates the application and controller layers, allowing for additional flexibility and edits to the network in real time.
Network intelligence is managed in a single centralized location where devices check in for the latest updates, instructions, and configurations. With the flexibility of virtualization and software much if not all of the SDN can be vendor agnostic, simplifying design and creating a more dynamic ecosystem.
Intent-Based Networking Systems (IBNS)
Intent-based networking systems use AI to optimize the network for a human-specific intent. Instead of coding, network administrators define an intent as an outcome or business objective. Artificial intelligence (AI) uses machine learning (ML) to optimize the network for that intent by establishing routines, setting policies, and responding to system events.
IBNS is seen as the most advanced form of orchestration as it does much of the technical work on its own, and allows humans to tweak the ML algorithms to perform more efficiently. Much of the manual work in IBNS is done during the training of the ML algorithm, and during the translation of intent.
Instructions are provided to the IBNS and those are interpreted by the algorithm as the networks overall goals. This could be a specific service level agreement (SLA) thresholds for device groups and applications - such is the case for Celona's unique MicroSlicing technology - or certain QoS rules that the organization wishes to always maintain.
After approval, the AI deploys the changes it thinks will match the intent of the instructions and creates a feedback loop to audit the network overtime. Over time, data will be made available to both the ML algorithm and administrators to measure how effective the intent-based network is at any given set of instructions.
While this is arguably the most effective way to deploy orchestration at scale, it solely relies on the speed of which machine learning algorithms achieve new milestones, which can take significant amounts of time depending on the intent.
Network orchestration use cases
Network orchestration can offer many benefits, including:
- Ensuring QoS policies are being followed
- Automating troubleshooting
- Auto configuration of new devices
- Establishing overlays for the control plane and the forwarding plane
- Provisioning network services
- Providing workflow automations
Orchestration doesn’t just empower enterprise Wi-Fi networks, but private LTE and 5G networks too.
Private Mobile Network Orchestration and Celona
Network orchestration can help enable networks to optimize, perform, and scale. So while orchestration helps with small-scale local area networks, it’s critical for large-scale enterprise networks, such as private mobile networks where organizations require top speed, coverage, and reliability. With the accelerating adoption of private 5G solutions such as Celona's, network orchestration is now more important than ever to optimize network performance without direct human mediation.
Mobile Network Operators (MNOs) use network orchestration to design private networks around certain requirements, monitor and optimize connections across the Radio Access Network (RAN), and configure new devices automatically upon joining the network.
Network slicing and QoS are also a component of orchestration, and rely on many self-learning technologies in order to work without intervention from administrators. QoS requirements can be entered into the edge device to essentially ‘train’ the algorithm on what devices and services to prioritize first.
Many elements of orchestration in the cellular world revolve around Self-Organizing Networks, or SONs. SONs play an important role in network automation as these functions allow for self-optimization, automation configuration, and self-healing based on real time analytics.
In the past many of these functions would have needed to be carried out manually, or through rigid pre-programmed scripts. With network orchestration in place, SONs and AI-powered algorithms can scan the network for any new changes, and modify their actions based on the best possible option for the current status of the network or device.
With the advent of private spectrum for the use of LTE and 5G wireless in the enterprise, such as CBRS in the United States, network orchestration technology is now available to private enterprises utilizing their own mobile networks. Celona takes advantage of all three network orchestration techniques to remove significant amount of friction in simplifying adoption of private cellular within enterprises.
This advancement in cellular technology will help pave the way for private 5G networks supporting large scale operations such as autonomous mobile robots and guided vehicles (AMR / AGV), realtime decision making with high density of IoT infrastructure, and mission critical voice and video communications.
Want to learn more about network orchestration? Subscribe to our newsletter below for the latest breakthroughs, and check out our Journey page to see the Celona solution in action.