Graph-Based Infrastructure Configuration for Scalable Change Validation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network management tools struggle to efficiently manage the growing complexity of enterprise data center infrastructures due to diverse network topologies and increasing data center sizes, lacking scalability and flexibility in adapting to changing requirements, leading to inefficient and costly reconfiguration processes.
Innovation Solution
A graph representation data model is utilized to configure and manage computing infrastructure, allowing for modular and scalable management by dividing processing into decentralized agents associated with specific graph representation patterns, enabling incremental updates and real-time state-based communication.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional network management tools are used to manage growing complexity of enterprise data center infrastructures, then basic configuration tasks can be performed, but scalability and flexibility are insufficient leading to inefficient and costly reconfiguration processes
Solution Approach 1:
The patent segments the network management system into multiple independent agents (configuration agents, validation agents, telemetry agents) that operate autonomously on different network devices. Each agent handles specific configuration tasks locally, eliminating the need for centralized reconfiguration processes and enabling independent modification of individual components without affecting the entire system.
Solution Approach 2:
The system implements dynamic configuration management where agents can independently update their local configurations based on changing requirements. The declarative intent specifications allow the system to adapt dynamically to new requirements without manual reconfiguration, and agents automatically propagate changes only where needed rather than reconfiguring entire infrastructures.
2Manufacturing precision
If declarative requirements specify very specific types of requirements for a particular type of network architecture, then the network configuration can be precisely controlled, but flexibility to utilize different network architectures is lost
Solution Approach 1:
The patent implements universal intent specifications that define desired network behavior and outcomes without being tied to specific architecture types. Agents are designed to be architecture-agnostic, capable of interpreting declarative requirements and implementing them across diverse network topologies (SDN, overlay, underlay, hybrid) using appropriate protocols and mechanisms for each architecture type.
Solution Approach 2:
The system separates the abstract intent parameters (service level agreements, performance requirements, security policies) from the concrete implementation parameters (specific protocols, device types, topology configurations). This allows the same declarative requirements to be satisfied through different parameter combinations depending on the chosen network architecture, enabling precise control over service outcomes while maintaining flexibility in implementation choices.
3Stability of the object's composition
If the entire computationally costly pipelined process is performed again in its entirety to implement minor changes, then configuration consistency is maintained, but computational overhead and time are excessive
Solution Approach 1:
The patent extracts the configuration validation and propagation logic from a monolithic centralized process and distributes it across multiple independent agents. When a minor change is needed, only the specific agent and its associated validation rules are executed, rather than re-running the entire pipelined process. This maintains configuration consistency through localized validation while dramatically reducing computational overhead.
Solution Approach 2:
Agents perform preliminary local validation of configuration changes before propagation, using pre-defined validation rules and constraints stored locally. This preliminary action ensures consistency is maintained at the source of changes, preventing invalid configurations from spreading through the system and eliminating the need for subsequent full-system validation passes.
4Reliability
If centralized management approaches are used, then overall network control is maintained, but scalability to large data center sizes is limited
Solution Approach 1:
The patent segments network management control functions into distributed agents deployed across multiple network devices. Each agent maintains local control over its device while adhering to declarative intent specifications. This segmentation enables scalability to large data center sizes as no single centralized point needs to manage all devices, yet configuration consistency is maintained through shared intent specifications and inter-agent communication.
Solution Approach 2:
Declarative intent specifications serve as intermediaries between centralized policy requirements and distributed device configurations. The intent specifications translate high-level management requirements into device-specific configuration parameters that distributed agents can independently implement. This intermediary layer maintains control over network configuration while enabling scalable distributed management architecture.
Data Source
Figure 1
Figure 2
Figure 3A
AI summary
A method for configuring a computing infrastructure is disclosed. The method comprises representing at least a portion of the computing infrastructure as a graph representation of computing infrastructure elements including a computing infrastructure node and a computing infrastructure edge, detecting a change in the graph representation of computing infrastructure elements, and determining whether the change affects a graph representation query pattern. In the event the change affects the graph representation query pattern, the change is notified to a query agent associated with the graph representation query pattern.