Telemetry Aggregation for Network Autonomic Scaling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Large-scale networks face challenges in dynamically monitoring and optimizing traffic engineered (TE) tunnels and virtual networks to meet service level agreements (SLAs) due to the generation of large quantities of raw performance information, lacking mechanisms to satisfy performance parameters beyond service availability and bit error rate.
Innovation Solution
Implementing telemetry-based data models that facilitate the aggregation and abstraction of performance telemetry data, allowing clients to subscribe to key performance monitoring and enabling autonomic network scaling capabilities, which include sending traffic monitor requests, receiving reports, and initiating scaling operations based on predetermined conditions such as bandwidth thresholds.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If large-scale networks monitor and optimize traffic engineered tunnels and virtual networks to meet service level agreements, then service quality and reliability are improved, but the system generates large quantities of raw performance information that becomes difficult to manage and process
Solution Approach 1:
The patent extracts only the essential performance parameters from the large volume of raw performance information. Network elements monitor multiple metrics but only report aggregated performance data that violates SLAs or requires attention, filtering out normal operational data. This extraction principle reduces the information management burden while maintaining reliability monitoring.
Solution Approach 2:
The patent introduces network controllers as intermediary components between network elements and management systems. These controllers aggregate performance data from multiple network elements, process the raw information locally, and only forward relevant summaries to central management. This intermediary layer prevents information overload at the management level while maintaining comprehensive monitoring capability.
2Measurement precision
If the network monitors comprehensive performance parameters beyond basic availability and bit error rate, then service quality measurement is improved, but the complexity of monitoring mechanisms increases
Solution Approach 1:
The patent segments the monitoring function across multiple hierarchical levels: network elements perform local metric collection, network controllers perform regional aggregation and analysis, and management systems perform overall SLA compliance verification. This segmentation distributes complexity across the architecture rather than concentrating it in a single complex monitoring system.
Solution Approach 2:
The patent enables network elements to autonomously monitor their own performance metrics and self-report when SLA violations occur. Each network element maintains local state information and independently determines when external monitoring intervention is needed, reducing the complexity of centralized monitoring mechanisms while achieving comprehensive performance measurement.
Data Source
AI summary
A method implemented by a central controller is provided. The method includes receiving traffic monitoring information from a client, and sending a traffic monitor request to a network controller, the traffic monitor request requesting the network controller to monitor one or more traffic parameters based on the traffic monitoring information. The method further includes receiving a traffic report from the network controller, and responsive to determining that a monitored traffic parameter in the traffic report does not satisfy a condition specified by the traffic monitoring information, initiating a scaling operation based on data contained in the traffic report.