CSPs deliver certain services that are considered critical to infrastructure and are therefore tightly regulated by many governments. These regulations force a level of service quality to which over-the-top (OTT) providers do not adhere. CSPs must conform to specific service quality metrics, for example resolving service disruptions quickly and automatically without packet loss affecting service quality. The same applies to services offered from a CSP to enterprise customers. Service quality is at the core of brand protection and customer experience management. As such, SLAs require the delivery of carrier grade quality services.
The following examples are essential NFV platform requirements for carrier grade systems:
- High Availability and Resiliency. The platform must provide integrated high availability and fault tolerance across the NFVI, virtual, and management domains. In the event of a failure, the platform must be able to self-heal to maximize service uptime. Mean Time To Failure (MTTF) must increase over the lifetime of the platform through adaptive, proactive, and automated monitoring systems. Mean Time To Repair (MTTR) must decrease over the lifetime of the NFV environment, as the system is optimized and proactive alerting takes place.
- Performance. The platform must achieve deterministic performance. The amount of resources required to deliver a certain level of performance must be well understood. Data plane intensive VNFs must be supported by the same components as control and management plane VNFs.
- Scalability. CSPs expect growth in the number of customers and services deployed on the NFV platform. The platform must be able to provide long term scale out capabilities, and dynamic and short term scale up and scale out functions.
- NFVI Life Cycle (Patching and Upgrades). The platform must be patched and upgraded by using optimized change management approaches for zero to minimal downtime.