Proxy Baseline Network Element Clustering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In complex data networks, calculating baselines for every monitored network element is resource-intensive and time-consuming, especially when gauging 'normal' behavior for millions of elements, making it impractical to perform individual baseline calculations for each performance metric across numerous network devices.
Innovation Solution
The method involves clustering network elements using k-means clustering to identify proxy elements, calculating proxy baselines for these elements, and using those baselines to monitor the performance metrics for the entire cluster, reducing the need for extensive baseline calculations across all network elements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If baseline calculations are performed for every monitored network element, then measurement precision is improved, but productivity deteriorates due to significant processing overhead and time consumption
Solution Approach 1:
The patent segments the network elements into clusters based on their performance characteristics and relationships. Instead of calculating baselines for every individual network element, the system divides them into groups where elements within the same cluster share similar baseline patterns. This segmentation reduces the total number of baseline calculations required while maintaining measurement precision through the use of proxy baselines for each cluster.
Solution Approach 2:
The patent creates proxy baselines that represent and copy the baseline behavior of multiple network elements within a cluster. Rather than performing individual baseline calculations for each element, the system generates a representative proxy baseline for the cluster that can be used to monitor all elements in that cluster. This copying approach significantly reduces processing overhead while preserving the essential baseline information needed for performance monitoring.
2Measurement precision
If baseline calculations are performed for every monitored network element, then measurement precision is improved, but loss of time increases due to significant processing overhead
Solution Approach 1:
The patent segments the network elements into clusters based on their performance characteristics and relationships. Instead of calculating baselines for every individual network element, the system divides them into groups where elements within the same cluster share similar baseline patterns. This segmentation reduces the total number of baseline calculations required while maintaining measurement precision through the use of proxy baselines for each cluster.
Solution Approach 2:
The patent performs preliminary clustering of network elements based on their performance characteristics and relationships before baseline calculation. By pre-grouping elements into clusters with similar behaviors, the system prepares the data structure in advance to enable efficient proxy baseline generation. This preliminary action reduces the time required for subsequent baseline calculations and performance monitoring operations.
3Measurement precision
If individual baseline calculations are performed for each network element, then measurement precision is improved, but device complexity increases due to memory and disc space requirements
Solution Approach 1:
The patent segments the network elements into clusters based on their performance characteristics and relationships. Instead of calculating baselines for every individual network element, the system divides them into groups where elements within the same cluster share similar baseline patterns. This segmentation reduces the total number of baseline calculations required while maintaining measurement precision through the use of proxy baselines for each cluster.
Solution Approach 2:
The patent merges multiple network elements into clusters and combines their baseline requirements into a single proxy baseline for each cluster. By combining the baseline monitoring needs of multiple elements into one representative baseline, the system significantly reduces memory and disc space requirements. The proxy baseline serves as a consolidated representation that captures the essential performance characteristics of all elements in the cluster.
Data Source
AI summary
A method may provide performance management for a data communication network including a plurality of network elements. The method may include defining a cluster of the network elements for a performance metric, and defining one of the network elements of the cluster as a proxy network element for the cluster. A proxy baseline of the performance metric for the cluster may be calculated based on performance metric data for the proxy network element, and the performance metric for each of the network elements of the cluster may be monitored using the proxy baseline of the performance metric. Related systems and computer program products are also discussed.