Distributed Primary Processing for Network Performance Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing communication network performance management systems face challenges in efficiently handling large amounts of traffic measurement data, particularly in complex networks with many small network elements, leading to capacity limitations and high resource usage, especially when real-time data collection and processing are required.
Innovation Solution
A distributed primary processing architecture is implemented, where both managing and managed systems have primary processing means, with processing control mechanisms to allocate data processing based on policy rules, allowing for flexible distribution of measurement data processing between the managing system and managed systems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all measurement data is collected and transferred to the managing system for processing, then centralized processing ensures uniform management, but transmission capacity is exceeded and network load increases
Solution Approach 1:
The patent divides the centralized processing function into distributed processing units located at different network elements. Each processing unit handles measurement data locally, segmenting the overall processing load across multiple independent units rather than concentrating it at a single managing system, thereby reducing transmission requirements while maintaining management uniformity through standardized interfaces
Solution Approach 2:
The patent introduces a new dimensional approach by implementing processing capabilities at the network element level (distributed dimension) rather than solely at the managing system level (centralized dimension). This creates a multi-dimensional processing architecture where data can be processed both locally and centrally, transforming the single-point processing model into a distributed networked model
2Loss of time
If frequent data collection is implemented to meet real-time response requirements, then response time is improved, but network capacity is exceeded and processing load increases
Solution Approach 1:
The patent implements preliminary processing of measurement data at the network element level before transfer to the managing system. By performing initial filtering, aggregation, and preprocessing operations locally, the system reduces the volume of data requiring frequent transmission, thereby achieving real-time responsiveness without proportionally increasing network capacity requirements
Solution Approach 2:
The patent extracts essential processing functions from the managing system and places them at the network element level. This extraction of processing capabilities allows local elements to independently handle time-critical operations, reducing the frequency and volume of data transfers required to meet real-time response requirements
3Measurement precision
If complex aggregation measurements are performed in the managing system, then comprehensive analysis is achieved, but processing time and resource consumption increase
Solution Approach 1:
The patent segments the aggregation measurement process into multiple stages: local filtering and preliminary aggregation at network elements, intermediate aggregation at regional collectors, and final comprehensive analysis at the managing system. This multi-level segmentation reduces the time required for complex measurements by distributing computational tasks across multiple hierarchical levels rather than performing all operations centrally
Data Source
AI summary
The present invention relates to an arrangement for performance management in a communication network comprising a managing system and a number of managed systems. The arrangement comprises collecting means (31) for collecting traffic measurement data and primary processing means for primary processing of measurement data. Said primary processing means are adapted to be distributed and comprise first primary processing means (11) provided in the managing system and a number of second primary processing means (21) provided in a number of managed systems. The arrangement also comprises processing control means (12,22) for controlling at least allocation of primary processing of measurement data to a first (11) or to a second (22) primary processing means.