Predictive Network Control for SLA Enforcement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current network optimization techniques rely on reactive methods, which are complex and difficult to manage due to the dynamic nature of traffic and network conditions, making it challenging to distinguish and optimize business and non-business critical traffic flows effectively, especially with increasing service level agreements (SLAs) and diverse application requirements.
Innovation Solution
A predictive networking approach using machine learning techniques to evaluate future network requirements and performance, allowing for proactive adjustments to ensure SLAs are met, implemented through a dynamic architecture that includes Traffic Pattern Analyzer (TPA), Network Analytics Manager (NAM), and Predictive Control Manager (PCM), which utilize learning machines to track and predict traffic and network attributes, and perform closed-loop control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If reactive networking techniques are used to enforce SLAs, then network performance criteria can be monitored and adjusted, but the configuration becomes extremely complicated and difficult to manage
Solution Approach 1:
The patent applies preliminary action by using machine learning models to predict future network performance and proactively adjust parameters before SLA violations occur. The system continuously trains models on historical data and uses them to forecast future states, enabling preventive rather than reactive network management. This reduces the complexity of continuous monitoring and adjustment by automating predictions and pre-computing optimal parameters.
Solution Approach 2:
The patent implements self-service through autonomous machine learning systems that automatically adjust network parameters without requiring complex manual configuration. The system self-trains on network data, self-evaluates performance, and self-adjusts parameters to meet SLAs. This eliminates the need for complicated manual configuration and reduces operational complexity while maintaining reliable SLA enforcement.
2Ease of operation
If traditional reactive networking mechanisms are used, then traffic classification and QoS can be controlled, but the system becomes highly complex due to dynamic traffic nature
Solution Approach 1:
The patent replaces traditional mechanical reactive control mechanisms with machine learning-based predictive control. Instead of using complex real-time classification and reactive adjustment systems, the patent uses trained ML models to predict traffic patterns and pre-determine optimal QoS parameters. This substitution of mechanical control with intelligent prediction simplifies the system while maintaining effective traffic control capability.
Solution Approach 2:
The system performs preliminary action by pre-training machine learning models on historical traffic data and pre-computing optimal QoS parameters before actual traffic control is needed. This allows the system to handle dynamic traffic patterns with simpler, pre-determined rules rather than complex real-time classification algorithms, reducing overall system complexity while preserving traffic control effectiveness.
3Adaptability or versatility
If multiple SLAs are managed with reactive techniques, then diverse application requirements can be addressed, but the management task becomes extremely complicated
Solution Approach 1:
The patent applies self-service by implementing autonomous machine learning systems that automatically manage multiple SLAs without requiring complex manual intervention. The system self-adapts to different application requirements by continuously learning from network data and automatically adjusting parameters for each SLA. This eliminates the complexity of manual multi-SLA management while maintaining the ability to address diverse application requirements effectively.
Solution Approach 2:
The patent uses parameter changes by dynamically adjusting QoS parameters based on predicted performance and trained models. Instead of manually configuring each SLA with fixed parameters, the system automatically modifies parameters in real-time based on ML predictions and current network conditions. This automated parameter adjustment simplifies management while enabling flexible adaptation to multiple SLA requirements.
Data Source
AI summary
In one embodiment, a management system determines respective capability information of machine learning systems, the capability information including at least an action the respective machine learning system is configured to perform. The management system receives, for each of the machine learning systems, respective performance scoring information associated with the respective action, and computes a degree of freedom for each machine learning system to perform the respective action based on the performance scoring information. Accordingly, the management system then specifies the respective degree of freedom to the machine learning systems. In one embodiment, the management system comprises a management device that computes a respective trust level for the machine learning systems based on receiving the respective performance scoring feedback, and a policy engine that computes the degree of freedom based on receiving the trust level. In further embodiments, the machine learning system performs the action based on the degree of freedom.