Stochastic Cross-Layer Network Optimization for Cost Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Large-scale network designs often result in over-provisioning, leading to significant costs due to manual optimization based on user experience and judgment, which fails to efficiently meet traffic demands while minimizing costs.
Innovation Solution
A method and system for generating a minimum monetary cost network model that iteratively updates network configurations to satisfy traffic demands by optimizing network failures, using a cross-layer optimization approach that considers physical and logical layer features, costs, and constraints to reduce network expenses.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual optimization based on user experience and judgment is used to design networks, then network designs can be created, but over-provisioning occurs leading to significant costs
Solution Approach 1:
The patent transforms the network design problem from manual parameter adjustment to automated optimization by changing the parameter representation to mathematical models. It uses stochastic programming parameters to represent uncertainty in traffic demands and failure scenarios, allowing the system to find optimal provisioning levels that meet availability targets without over-provisioning.
Solution Approach 2:
The patent replaces the mechanical/manual network design process with an automated computational optimization system. It substitutes human judgment and experience with stochastic cross-layer optimization algorithms that systematically evaluate multiple scenarios and automatically determine optimal network configurations that minimize costs while meeting reliability targets.
2Reliability
If over-provisioning is used to meet traffic demands, then traffic flow availability is improved, but network costs increase significantly
Solution Approach 1:
The patent applies partial action by provisioning exactly the amount of network capacity needed to meet availability targets under various failure scenarios, rather than excessive provisioning. It uses stochastic optimization to determine the precise provisioning level that satisfies reliability requirements without unnecessary over-provisioning, thereby reducing network costs while maintaining adequate service levels.
Solution Approach 2:
The patent changes the approach from fixed provisioning to dynamic, scenario-based provisioning parameters. It uses stochastic programming to model different traffic demand scenarios and failure conditions, allowing the network design to adapt provisioning levels to actual needs rather than assuming worst-case scenarios for all situations.
3Adaptability or versatility
If separate optimization is performed for each service objective, then each objective can be addressed, but the overall network design becomes complex and inefficient
Solution Approach 1:
The patent merges multiple separate service objective optimizations into a unified cross-layer stochastic optimization framework. It combines physical layer, data link layer, and network layer considerations into a single integrated model that simultaneously optimizes for traffic flow availability, cost, and reliability across all service objectives, reducing overall design complexity while maintaining comprehensive coverage.
Solution Approach 2:
The patent creates a universal optimization framework that handles multiple service objectives through a single stochastic cross-layer model. This multi-functional approach allows the same optimization engine to address different service requirements (traffic engineering, reliability, cost optimization) simultaneously, rather than requiring separate optimization processes for each objective.
Data Source
AI summary
The present disclosure describes system and methods for network planning. The systems and methods can incorporate network traffic demands, availability requirements, latency, physical infrastructure and networking device capability, and detailed cost structures to calculate a network design with minimum or reduced cost compared to conventional methods. In some implementations, the method include providing an initial, deterministic set of failures, and then successively performing a network optimization and a network availability simulation to determine which failures most impact the performance of the network model. The high impact failures can then be provided back into the system, which generates an improved network design while still maintaining minimum cost.