Load Balanced Ethernet Over SONET Network Architecture
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network architectures, both circuit-switched and packet-switched, face challenges in scaling to accommodate large mesh topologies and suffer from bandwidth inefficiency and high packet switching capacities, particularly in handling bursty data traffic.
Innovation Solution
A load-balanced network architecture that splits incoming traffic flows into equal parts and distributes them across multiple nodes, using pre-provisioned circuits to route them to destination nodes based on packet headers, allowing for efficient bandwidth utilization without requiring high packet switching capacities at each node.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If circuit-switched network architecture is used, then reliability and protection mechanisms are improved, but bandwidth efficiency deteriorates
Solution Approach 1:
The patent segments a single traffic flow into multiple parts and distributes them across different pre-provisioned circuits. This allows circuit-switched networks to handle bursty data traffic more efficiently by utilizing statistical multiplexing across multiple fixed-bandwidth connections, thereby improving bandwidth efficiency while maintaining the reliability benefits of circuit switching.
2Loss of energy
If packet-switched network architecture is used, then bandwidth efficiency is improved, but packet switching capacity requirements deteriorate
Solution Approach 1:
By segmenting traffic flows into multiple parts and distributing them across pre-provisioned circuits, the patent reduces the packet switching capacity required at each individual node. Each node only needs to handle a fraction of the total traffic, lowering the switching capacity requirements while maintaining bandwidth efficiency through statistical multiplexing.
Solution Approach 2:
The patent introduces pre-provisioned circuits as intermediaries between source and destination nodes. These circuits handle the actual traffic transmission, reducing the burden on packet switching capacity at intermediate nodes while maintaining the benefits of packet switching for handling bursty traffic.
3Productivity
If network architecture is scaled to accommodate large mesh topologies, then network capacity is improved, but switching complexity deteriorates
Solution Approach 1:
The patent segments traffic flows and distributes them across multiple pre-provisioned circuits, which simplifies the switching logic at each node. Instead of requiring complex dynamic routing decisions for large mesh topologies, nodes use pre-established circuits to forward segmented traffic, reducing switching complexity while maintaining network capacity.
Solution Approach 2:
The patent uses pre-provisioned circuits that are established in advance before traffic flows are routed. This preliminary action eliminates the need for complex real-time routing decisions at each node, significantly reducing switching complexity while allowing the network to scale to large mesh topologies.
Data Source
AI summary
A load-balanced network architecture is disclosed in which a traffic flow at a given network node is split into a plurality of parts, and the parts are distributed to respective ones of the plurality of nodes that are designated as participating in a load balancing process for the traffic flow. Each of at least a subset of the participating nodes receiving one of the parts routes at least a portion of its received part to one or more destination nodes.