Private Network Event Prioritization for Demand Surge Orchestration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional network management methods struggle to effectively anticipate and manage transient, localized surges in demand on private networks, leading to resource strain and congestion due to a diverse user mix with varying connectivity requirements and service level agreements, failing to rapidly scale up or down infrastructure while maintaining alignment with pre-existing policies.
Innovation Solution
A management system utilizing machine learning models to detect, prioritize, and dynamically orchestrate network resources for strategic traffic steering and load balancing, aligning with real-time fluctuations and user demands to conserve resources and maintain network integrity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional network management methods are used, then network infrastructure can handle steady-state traffic, but it fails to effectively anticipate and manage transient, localized surges in demand leading to resource strain and congestion
Solution Approach 1:
The system performs preliminary actions by detecting events and predicting future network demand before the surge occurs. The machine learning models analyze current network state and event data to forecast upcoming demand patterns, allowing the system to proactively allocate resources and reconfigure network infrastructure in advance of actual demand spikes, thereby preventing congestion and maintaining reliable performance.
2Adaptability or versatility
If network resources are allocated to handle diverse user mix, then connectivity requirements can be met, but computing and networking resources are consumed inefficiently
Solution Approach 1:
The system implements dynamic resource allocation that adapts to diverse user connectivity requirements in real-time. Instead of statically provisioning resources for all possible user scenarios, the machine learning models continuously analyze network state and event data to dynamically adjust resource allocation, scaling up only when and where needed to meet specific user demands while conserving computing and networking resources during normal operation.
3Reliability
If network infrastructure is scaled up to handle peak demand, then service level agreements can be maintained, but resources are wasted during low-demand periods
Solution Approach 1:
The system uses preliminary prediction to maintain SLA compliance without over-provisioning. Machine learning models forecast upcoming demand surges based on event data and network state analysis, enabling the system to scale resources up just before peak demand occurs and scale them down afterward. This predictive approach ensures SLA compliance during critical periods while maximizing resource utilization efficiency during low-demand periods.
4Ease of operation
If traditional methods manage network demand, then basic connectivity is provided, but transient localized demand surges cause congestion and resource strain
Solution Approach 1:
The system implements continuous feedback loops where machine learning models monitor network state, event data, and performance metrics in real-time. This feedback enables the system to detect emerging demand patterns, predict upcoming surges, and automatically adjust resource allocation and traffic routing to prevent congestion before it occurs, thereby maintaining both ease of operation and network stability during events.
Data Source
AI summary
A device may detect events associated with users of a public network and a private network, and may process the events, with a first machine learning model, to define the events and assign priorities to the events to generate prioritized events. The device may process the prioritized events and policies associated with the private network and the public network, based on the priorities and with a second machine learning model, to determine actions to perform in the private network and the public network. The device may cause resources to be allocated in the private network and the public network for performance of the actions.