PCECC Distributed Path Computation for Network Overload
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Network Path Computation Elements (PCEs) become overloaded when handling numerous path computation requests, leading to degraded performance due to manual node assignment and complexity in communicating across multiple protocols, resulting in inefficiencies and increased costs.
Innovation Solution
A PCE Central Controller (PCECC) distributes path computation tasks to dual-configured Path Computation Clients (PCCs), enabling the determination of network slices and assignment of nodes across multiple domains using the Path Computation Element Communication Protocol (PCEP), thereby optimizing path computation and reducing overload.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a single PCE handles all path computation requests, then comprehensive path computation capability is achieved, but the PCE becomes overloaded and performance degrades
Solution Approach 1:
The patent segments the centralized PCE into multiple distributed PCEs across different network domains. Each PCE handles path computation requests for its local domain, dividing the overall computation workload into smaller, manageable segments that can be processed independently, thereby preventing any single PCE from becoming overloaded while maintaining comprehensive path computation capability across the entire network.
2Ease of manufacture
If manual node assignment is used for PCE configuration, then implementation simplicity is maintained, but operational complexity increases and efficiency decreases
Solution Approach 1:
The patent implements self-service through automated node assignment mechanisms where PCEs automatically discover their roles and configurations based on network topology and demand patterns. The system autonomously assigns computational tasks to appropriate PCEs without manual intervention, and dynamically adjusts configurations based on real-time performance metrics, thereby maintaining implementation simplicity while dramatically improving operational efficiency.
3Adaptability or versatility
If multiple protocols are supported for inter-domain communication, then network compatibility is improved, but system complexity and communication overhead increase
Solution Approach 1:
The patent introduces standardized interface protocols as intermediaries between diverse network domains. These interface protocols act as mediators that translate between different domain-specific protocols and a common computation framework, enabling multi-protocol support and network compatibility while abstracting away the underlying complexity from the path computation logic.
4Measurement precision
If centralized path computation is used, then optimal path determination is achieved, but communication overhead and processing time increase
Solution Approach 1:
The patent segments the centralized computation process into distributed computational units that operate autonomously in their respective domains. Each segment performs local path optimization computations independently, eliminating the need for extensive inter-domain communication and data aggregation, thereby maintaining optimal path determination accuracy while significantly reducing computation time and communication overhead.
Solution Approach 2:
The patent implements preliminary action by pre-computing and caching path information for commonly traversed network segments. PCEs prepare and store optimization data in advance based on historical traffic patterns and network topology, so that when path computation requests arrive, the system can quickly retrieve and combine pre-computed segments rather than performing complete optimization calculations from scratch, reducing processing time while maintaining accuracy.
Data Source
Figure 1~2
Figure 3~4
Figure 5
AI summary
A path computation element (PCE) central controller (PCECC) comprising a memory comprising executable instructions and a processor coupled to the memory and configured to execute the instructions. Executing the instructions causes the processor to receive a request to compute a path through a network, the request comprising a plurality of computational tasks, divide the computational tasks into a plurality of groups of computational tasks, transmit at least some of the plurality of groups of computational tasks to a plurality of path computation clients (PCCs) for computation by the PCCs, and receive, from the PCCs, computation results corresponding to the plurality of groups of computational tasks.