Benefits

Use the orchestration component of IBM® Cloud Pak for Network Automation to design and integrate external resources into virtual production environments and then automate the management of end-to-end lifecycle processes. This approach is known as network function virtualization. Read about the benefits of this approach and the key functions in the orchestration component, and also view a case study.

Benefits of network function virtualization (NFV)

NFV brings a significant operational paradigm shift for service providers. Today’s physical network appliances require highly manual processes to manage their end-to-end lifecycle. Testing, installation, configuration, and problem management of network appliances all revolve around manual activities that often require a physical truck roll or a human to run each lifecycle process.

NFV’s software paradigm promises fully automated lifecycle processes for bringing network services into production and maintaining them thereafter. Virtual network functions (VNF) allow a much simpler set of lifecycle tasks enabling near full automation of the creation and healing of virtual services, far more than is possible with their physical counterparts.

Vast business opportunities are associated with NFV transformation, including new revenue streams, improved customer experience, and reductions in both operational and capital expenditure. However, a fully automated lifecycle solution for NFV comes with extra complexities.

The orchestration component of IBM Cloud Pak for Network Automation is a comprehensive services design, testing, and automated deployment platform that addresses the challenges and complexities of the NFV paradigm. It delivers an end-to-end automated service lifecycle solution from initial design to production, as depicted in the following figure:

Core features and benefits

The key differentiating features of the orchestration component that is shown in this figure include:
  • A common way of handling resources through a unified lifecycle model
  • Support for quick resources and assembly onboarding
  • Intent-driven lifecycle management
  • Support for quality management and policy
  • Cloud-native solution

The benefits of using the orchestration component of IBM Cloud Pak for Network Automation to deliver NFV are illustrated in the following figure:

Orchestration component benefits

Lifecycle management

To achieve NFV’s promised levels of automation, the orchestration component of IBM Cloud Pak for Network Automation provides a complete DevOps toolchain that manages the end-to-end lifecycle of virtual network services, from release management of VNF software packages to the continuous orchestration and running of VNFs and service instances.

Third-party VNF software must be wrapped in a well-tested standard lifecycle interface, and service bundles of multi-vendor VNFs must be tested for interoperability and performance to ensure that there are no errors. This ensures that in production, services and VNFs can be constantly created, configured, updated, scaled, healed, and migrated without manual intervention.

The complete lifecycle management is shown in the following figure:

The orchestration component's end-to-end DevOps toolchain

VNF and services onboarding requires a comprehensive release management strategy, a suite of tools and a lifecycle integration framework to accommodate the variety of third-party VNF software package formats. Continuous in-life orchestration also requires a new approach to modelling and managing the complexity of in-life network function lifecycle management. To achieve the levels of automation required, a much simpler and standardized approach is required to implement all foreseen lifecycle transitions.

Advantages of an end-to-end DevOps deployment model

The advantages of adopting an end-to-end DevOps deployment model are illustrated in the following figure:

Advantages of the end-to-end DevOps toolchain

Case study: Heal operation that uses orchestration

In this closed loop heal case study, a server no longer receives IP packets. An alarm is raised. The event is evaluated and a Heal event is triggered and run by the orchestration component. The solution that addresses this case study includes the following components (some of which might be included in a service assurance product):
  • A service level management (SLM) system
  • A policy manager
  • A topology manager
  • The orchestration component of IBM Cloud Pak for Network Automation

The heal solution process is described in the following steps:

  1. The SLM system, acting as an external operational support system (OSS), receives an alarm that a server has stopped receiving IP packets.
  2. This alarm then triggers a policy in the policy manager.
  3. As part of the service design and assembly onboarding, the topology manager reconciles the orchestration assemblies with OpenStack virtual machines.
  4. The policy manager (acting as operations) now has enough knowledge to trigger an orchestration Heal request.
  5. The orchestration component transitions the broken server through the following lifecycles:
    1. STOP
    2. START
    3. INTEGRITY
  6. The topology manager gets notified by the orchestration component that a lifecycle event has occurred on the broken server.
  7. This triggers the observer in the topology manager that raised the alarm to rerun.
  8. Rerunning the observer clears the SLM system alarm, confirming that the server is receiving packets again, and that healing was successful.