Programmable Path Computation for Bandwidth-Aware TE Routing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing IP routing technologies overutilize certain links and lack flow-by-flow control, leading to inefficient bandwidth management and limited control over traffic paths in network routing.
Innovation Solution
A programmable Path Computation Engine (PCE) that manages and calculates TE paths based on user-defined SR policies and TE tunnels, allowing for customizable rules and separate management of intent and path objects, with bandwidth tracking and rule-based engines to ensure efficient resource utilization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If IP routing is used for network traffic forwarding, then the routing process is simple and automatic, but all flows overuse certain low metric links leading to inefficient bandwidth management
Solution Approach 1:
The patent introduces a Path Computation Element (PCE) as an intermediary between the IP routing layer and traffic flows. The PCE receives path computation requests, calculates optimal paths considering bandwidth constraints, and returns computed paths to the requesting entities. This mediator enables flow-by-flow path control without requiring complex routing protocols throughout the entire network, thus improving bandwidth utilization efficiency while maintaining manageable system complexity.
Solution Approach 2:
The patent segments the routing control function into two parts: standard IP routing for basic forwarding and a separate path computation function for optimized routing. By dividing the routing control into these independent segments, the system can apply simple IP routing where standard paths suffice while using the PCE for specific flows requiring optimized paths, thereby improving overall bandwidth efficiency without making the entire routing system complex.
2Adaptability or versatility
If Traffic Engineering techniques are implemented to control paths flow-by-flow, then bandwidth resource control is improved, but the system complexity increases
Solution Approach 1:
The PCE serves as a centralized intermediary that handles all path computation tasks. Individual network elements don't need to implement complex path calculation algorithms; they simply request paths from the PCE and follow the computed paths. This concentrates the complexity in a single manageable entity while keeping individual network elements simple, thus providing flow-by-flow control flexibility without proportionally increasing overall system complexity.
Solution Approach 2:
The patent enables network elements to autonomously request and use computed paths without requiring complex coordination or negotiation protocols. Each element can independently initiate a path computation request to the PCE and implement the returned path, providing self-service capability that simplifies the system architecture while maintaining adaptability for different traffic engineering requirements.
3Productivity
If multiple SR policies and TE tunnels are managed together, then resource integration is improved, but management complexity increases
Solution Approach 1:
The patent separates SR policies and TE tunnels into distinct managed object sets. SR policies are managed as one type of object with their specific attributes and operations, while TE tunnels are managed as another type with their own attributes and operations. This segmentation allows each object type to be managed independently according to its specific requirements, improving resource management efficiency while avoiding the complexity of managing all objects uniformly.
Solution Approach 2:
The patent implements a universal management framework that can handle both SR policies and TE tunnels through common operations and interfaces. The management system provides unified functions for creating, modifying, and deleting both object types, as well as common operations like path computation and bandwidth tracking. This multi-functional approach improves resource integration and management efficiency while maintaining ease of operation through consistent management procedures.
Data Source
AI summary
A network control system is configured to manage controllable network elements or network paths through programmable rule-based engines operating at multiple scopes. A plurality of managed objects, including operator-intent objects such as Segment Routing (SR) policies and Traffic Engineering (TE) tunnels, and path-calculation objects such as managed paths, are maintained. Each rule-based engine is associated with a scope comprising at least an individual managed object scope, a set of related managed objects scope, and a network-wide scope. State information of the managed objects is monitored at each scope, and one or more customizable rules are applied to evaluate conditions based on the monitored state information. Responsive to the evaluations, the system performs fine-grained actions on individual managed objects and broad-scale actions across multiple managed objects within a common network control framework. Bandwidth tracking, constraint management, and coordinated group-level path recalculations are supported to optimize network performance and adapt to real-time events without code-level customization.