Data Network Server Hibernation via Traffic Profiling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data network management systems inefficiently identify and decommission underutilized servers, as they rely on manual configuration and unreliable metrics like data volume, which can lead to resource wastage and increased operational costs.
Innovation Solution
A method that monitors data flows, creates profiles of network traffic behavior, identifies low-usage servers, stores programming instructions and data, and shuts down these servers while allowing for their reinstatement as virtual servers, optimizing resource usage and reducing operational costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If simple policy trigger thresholds on data volume are used to initiate server shut down or hibernation, then the process is automated, but the reliability of identification is poor because data volume is an unreliable indicator of CPU utilisation
Solution Approach 1:
The patent changes the monitoring parameters from simple data volume metrics to comprehensive CPU utilisation metrics, including instruction count, cycle count, and other processor performance indicators. This allows reliable identification of actually unused servers while maintaining automated decision-making.
Solution Approach 2:
The patent introduces an intermediary analysis layer that correlates multiple metrics (data volume, CPU utilisation, instruction count, cycle count) to determine server usage status. This intermediary process reconciles the automated trigger mechanism with reliable identification by using composite indicators rather than single metrics.
2Loss of energy
If servers are shut down to save resources, then operational costs are reduced, but the risk increases that essential services may become unavailable if the server is needed
Solution Approach 1:
The patent performs preliminary actions by creating virtual copies and storing backup data before shutting down physical servers. This allows rapid reinstatement if needed, reducing the risk of service unavailability while maintaining cost savings from shutdown.
Solution Approach 2:
The patent creates virtual server copies that replicate the functionality of physical servers. These copies can take over immediately if the shutdown server is needed, ensuring service continuity while allowing the physical server to remain offline for resource savings.
3Measurement precision
If manual configuration is required to determine suitable trigger points for server shutdown, then the precision of identification can be improved, but the complexity and time required increases significantly
Solution Approach 1:
The patent enables the system to self-configure by automatically learning optimal trigger thresholds from historical data and usage patterns. This eliminates the need for manual configuration while maintaining high precision in identifying servers ready for shutdown.
Solution Approach 2:
The patent implements feedback mechanisms where the system continuously monitors server performance and usage patterns, automatically adjusting trigger thresholds based on observed behavior. This feedback loop maintains measurement precision without requiring manual intervention or complex configuration.
Data Source
Figure 1
Figure 2
AI summary
The servers in a data network are monitored (16) and classified (290) according to the data flows to and from the servers, to identify servers with low usage, and programming instructions and data relating to those servers are retrieved and stored (32, 17) so that they may subsequently be retrieved (18) to replicate the server to which they relate.