Network Device Trend Indicator Computation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for determining trend and tendency of state variables in network devices are inefficient, requiring repetitive polling and excessive resource usage, and do not provide on-demand information, leading to incomplete performance assessment.
Innovation Solution
Implementing a method within network devices to store and compute trend indicator and tendency values for state variables, allowing periodic sampling and immediate delivery of these values to management stations upon request, reducing the need for continuous monitoring and repetitive polling.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If repetitive polling is used to obtain trend information, then trend and tendency data can be obtained, but management application performance degrades and processing resources are excessively consumed
Solution Approach 1:
The network device pre-computes and stores trend indicator values and tendency values for state variables before any management application requests them. This preliminary computation eliminates the need for repetitive polling, as the device can immediately provide pre-calculated trend information when queried, thereby maintaining information availability while preserving management application performance.
Solution Approach 2:
The network device performs self-monitoring and self-computation of trend information by periodically sampling its own state variables and calculating trend indicators internally. This self-service approach transfers the computational burden from the management application to the network device itself, allowing the device to provide trend data on-demand without requiring continuous external polling.
2Loss of information
If historical values are stored in the device to enable trend analysis, then trend information becomes available, but device memory requirements and complexity increase
Solution Approach 1:
The patent extracts only the essential elements needed for trend analysis (trend indicator values and tendency values) from the complete historical data set. Instead of storing all historical state variable values, the device computes and retains only the derived trend metrics, significantly reducing memory requirements and device complexity while maintaining the ability to provide comprehensive trend information.
Solution Approach 2:
The invention transforms the raw state variable data into different parameter representations (trend indicator values and tendency values) that capture the essential trend information in a compressed form. This parameter transformation reduces the data volume that needs to be stored and processed, thereby lowering device complexity and memory requirements while preserving trend analysis capabilities.
3Loss of time
If continuous monitoring is implemented to provide real-time trend data, then up-to-date information is available, but resource consumption and system overhead increase
Solution Approach 1:
The network device implements periodic sampling of state variables at predetermined intervals rather than continuous monitoring. This periodic action maintains information freshness by regularly updating trend indicator values and tendency values, while significantly reducing processing resource consumption and energy usage compared to continuous monitoring approaches.
Data Source
AI summary
A method of providing performance trend information from a device in a packet-switched network to a network management station comprises maintaining in the device one or more state variables that store values associated with data processing functions of the device; storing in the device a trend indicator value and a trend tendency value, both associated with a particular state variable among the state variables; periodically sampling the particular state variable to obtain one or more sampled values; determining and updating the trend indicator value with a direction of change in the sampled values of the particular state variable over a particular time; determining and updating the trend tendency value with a rate of change in the one or more sampled values of the particular state variable over the particular time; and sending the trend indicator value and the trend tendency value to the network management station.