Hose-Based Network Capacity Planning for Traffic Uncertainty
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current network capacity planning methodologies are reactive and ad hoc, struggling with uncertainty in forecasting future network traffic growth, often relying on a single traffic matrix that is brittle and less effective in handling changes, leading to overprovisioning and inefficiencies.
Innovation Solution
The adoption of a hose-based approach in network planning, which abstracts aggregated traffic volume per site, allowing for multiplexing of peak traffic flows and buffering against spikes, combined with heuristic algorithms to select hose-compliant traffic matrices and cross-layer optimization between optical and Internet Protocol networks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single traffic matrix is used for network capacity planning, then the planning process is simplified, but the network becomes brittle and less effective at handling traffic changes and uncertainty
Solution Approach 1:
The patent transitions from static single traffic matrix planning to dynamic multi-scenario planning. Multiple traffic matrices representing different future scenarios (baseline, optimistic, pessimistic) are generated and used to create adaptive network capacity plans that can respond to actual traffic patterns, making the planning process dynamic rather than static.
Solution Approach 2:
The patent applies preliminary action by pre-generating multiple traffic matrices that represent potential future scenarios before actual network planning. This allows the network to be provisioned in advance for multiple possible outcomes, reducing the need for reactive changes when actual traffic patterns emerge.
2Reliability
If network capacity is overprovisioned to handle uncertainty, then network reliability improves, but resource efficiency and cost increase
Solution Approach 1:
The patent changes the parameter approach from single-point estimation to multi-point scenario analysis. By generating traffic matrices with different growth rates and patterns (baseline, optimistic, pessimistic scenarios), the system identifies capacity levels that satisfy multiple parameter sets, achieving reliability without excessive overprovisioning.
Solution Approach 2:
The patent applies partial action by provisioning capacity that satisfies the majority of scenarios rather than all possible scenarios. This avoids the excessive action of overprovisioning for every conceivable traffic pattern, achieving a balance between reliability and resource efficiency.
3Measurement precision
If traditional pipe-based models are used for network planning, then individual traffic flows are tracked precisely, but the system cannot effectively multiplex peak traffic flows or buffer against spikes
Solution Approach 1:
The patent merges individual pipe-based traffic flow tracking with aggregate flow-based planning. By combining detailed traffic matrix data with flow-based capacity allocation, the system maintains measurement precision for planning purposes while enabling multiplexing capabilities through aggregate resource management and statistical multiplexing of traffic flows.
Solution Approach 2:
The patent creates a universal planning framework that serves multiple functions: it tracks individual traffic flows for precision measurement, enables aggregate flow multiplexing for efficiency, and provides adaptability for handling traffic spikes. The multi-scenario traffic matrices make the planning system universally applicable to different traffic patterns and growth scenarios.
Data Source
AI summary
The disclosed computer-implemented method may include (i) generating a data center constraint model by placing a constraint on a total amount of ingress or egress traffic a service expects from each respective data center of multiple data centers, (ii) filtering a set of traffic matrices that indicate points in the data center constraint model by comparing the set of traffic matrices against cut sets of a network topology that indicate network failures to create a tractable set of dominating traffic matrices, (iii) obtaining physical network resources to implement a cross-layer network upgrade architecture that satisfies the tractable set of dominating traffic matrices, and (iv) allocating the physical network resources across the multiple data centers according to the cross-layer network upgrade architecture such that a capacity level of the multiple data centers is increased while satisfying the data center constraint model. Various other methods, systems, and computer-readable media are also disclosed.