Layered Availability Counting in NFV Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for calculating availability in network functions virtualization systems cannot determine availability at different layers, only providing overall system availability, which is insufficient for meeting high reliability requirements in telecommunications.
Innovation Solution
An availability counting apparatus and method that includes a configuration module to receive unavailability rules, a calculation module to process fault and recovery information from detection apparatuses at each layer, and a report module to output periodic availability reports, allowing for separate layer availability calculations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If overall system availability is counted using manual or third-party tools, then the availability calculation is simple, but availability of different layers of the system cannot be determined
Solution Approach 1:
The patent segments the NFV system into multiple layers (infrastructure layer, virtual network layer, and operation support layer) and implements separate availability detection and counting for each layer. The availability detection apparatus at each layer independently detects availability indicators specific to that layer, and the calculation module separately calculates availability metrics for each layer, enabling precise measurement of layer-specific availability while maintaining manageable system complexity through modular architecture.
2Reliability
If manual availability counting is performed, then device costs are reduced, but labor intensity and time consumption increase
Solution Approach 1:
The availability detection apparatus is integrated into the NFV system itself, enabling self-detection and self-reporting of availability status. Each layer's detection apparatus automatically detects its own availability indicators and reports them to the calculation module without requiring external manual intervention, thereby reducing labor intensity and time consumption while maintaining high reliability through automated, consistent measurement.
Solution Approach 2:
The system implements continuous feedback loops where availability indicators are detected, reported, and used to calculate availability metrics that are then output in periodic reports. This automated feedback mechanism eliminates manual counting delays and ensures timely, reliable availability assessment across all system layers.
3Reliability
If third-party tools are used for availability detection, then continuous connection monitoring is achieved, but system cost and external dependency increase
Solution Approach 1:
The patent merges the availability detection functionality directly into the NFV system architecture by integrating availability detection apparatus at each layer with the system's existing management and orchestration components. This eliminates the need for separate third-party tools, reducing external dependencies and simplifying system architecture while maintaining continuous monitoring capabilities through the unified integrated approach.
Data Source
Figure 1A
Figure 1B
Figure 2
AI summary
Embodiments of the present invention disclose an availability counting apparatus and method. The apparatus includes: a configuration module, configured to: receive an unavailability rule configured by a user, and subscribe to a detection item from a detection apparatus at each layer of network functions virtualization and notify the detection apparatus of a condition for triggering a fault alarm, according to the unavailability rule; a calculation module, configured to: receive fault alarm information and fault recovery information that are reported by the detection apparatus, and calculate availability of each layer of the network functions virtualization; and a report module, configured to output an availability counting report according to a preset period. According to the present invention, availability of layers of a network functions virtualization system can be counted separately.