Distributed Flow Control for TWIN Network Congestion

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

In time-domain wavelength interleaved networks, centralized scheduling faces challenges with unacceptably large propagation delays for asynchronously varying traffic, necessitating effective flow control and congestion management techniques to maintain network stability and fairness.

Innovation Solution

The implementation of distributed flow control algorithms that collect and analyze congestion information to adjust transmission rates, incorporating feedback mechanisms for dynamic scheduling and fairness, allowing nodes to operate under stable and fair transmission schedules, even with asynchronous data bursts.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Extent of automation

If centralized scheduling is used in TWIN architecture, then scheduling control can be achieved, but propagation delays become unacceptably large for asynchronously varying traffic

Engineering Contradiction:

Improvescheduling controlVSAvoidpropagation delays

Core Design Contradiction:

Extent of automationVSLoss of time

Solution Approach 1:

The patent segments the centralized scheduling function into distributed scheduling units at each network node. Each node independently makes scheduling decisions based on local congestion information and feedback from other nodes, eliminating the need for centralized control signals to propagate across the entire network and thus reducing propagation delays.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent implements feedback mechanisms where nodes exchange congestion information and scheduling status with neighboring nodes. This distributed feedback loop enables each node to adjust its scheduling decisions based on real-time network conditions, achieving coordinated scheduling without centralized control and minimizing propagation delays.

Inventive Principle:

Principle #23Feedback

2Speed

If distributed flow control algorithms are implemented to reduce propagation delays, then scheduling responsiveness improves, but system complexity increases

Engineering Contradiction:

Improvescheduling responsivenessVSAvoidsystem complexity

Core Design Contradiction:

SpeedVSDevice complexity

Solution Approach 1:

Each network node is equipped with autonomous scheduling capabilities that enable it to independently make scheduling decisions based on local congestion information. Nodes self-adjust their transmission rates and scheduling parameters without requiring complex centralized control logic, thereby improving responsiveness while keeping individual node complexity manageable.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent employs universal scheduling algorithms that can be implemented at each node with the same functional capabilities. Each node performs multiple functions including congestion detection, scheduling decision-making, and feedback generation using the same algorithmic framework, which simplifies the overall system architecture despite the distributed nature of the control.

Inventive Principle:

Principle #6Universality (Multi-functionality)

3Stability of the object's composition

If transmission rates are dynamically adjusted based on congestion feedback, then network stability and fairness are maintained, but control complexity increases

Engineering Contradiction:

Improvenetwork stabilityVSAvoidcontrol complexity

Core Design Contradiction:

Stability of the object's compositionVSDevice complexity

Solution Approach 1:

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

Nodes dynamically adjust transmission rate parameters based on congestion feedback while maintaining stable operation. The control complexity is managed by adjusting only the transmission rate parameter rather than reconfiguring the entire scheduling system, allowing for stable and fair network operation with controlled complexity.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS7573815B2Flow control and congestion management for random scheduling in time-domain wavelength interleaved networks

Publication Date: 2009.08.11 NOKIA OF AMERICA CORP

AI summary

Flow control techniques are disclosed for use in data communications networks such as those implementing a time-domain wavelength interleaved network (TWIN) architecture or other suitable architectures. Such techniques may provide for congestion management and scheduling for asynchronous traffic. For example, in one aspect, a technique comprises collecting information at a node of an optical-based communications network, wherein at least a portion of the collected information pertains to congestion in the optical-based communications network, and adjusting a transmission rate of the node in response to at least a portion of the collected information such that the node operates under a substantially stable transmission schedule with respect to asynchronous data bursts. The transmission rate adjusting step/operation may further comprise adjusting the transmission rate such that the node operates under a substantially fair transmission schedule. The node may operate under a random transmission schedule.