Multiple Shortest-Path Tree Protocol for Loop-Free Topology
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing computer networks often experience loop formation due to redundant communication paths, leading to circuitous packet traversal and network overload, which conventional spanning tree protocols fail to adequately address.
Innovation Solution
The Multiple Shortest-Path Tree (MSPT) protocol embeds control information for spanning tree creation within the source MAC address field of packets, allowing switches to calculate a loop-free network topology and populate forwarding tables efficiently, thereby avoiding loop formation and optimizing packet forwarding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If redundant communication paths are added to the network, then network reliability is improved, but loop formation occurs causing packets to traverse circuitous paths indefinitely
Solution Approach 1:
The patent segments the network topology into multiple shortest-path trees rooted at different network devices. Each tree represents a loop-free path structure, and the segmentation allows the network to maintain redundancy while avoiding loops by selecting appropriate trees for different traffic flows. This resolves the contradiction by dividing the complex loop-prone network into manageable loop-free segments.
Solution Approach 2:
The patent changes the parameter of path selection by using cost-based metrics to determine the shortest path in each spanning tree. By dynamically adjusting path costs and selecting optimal paths based on current network conditions, the system maintains reliability through redundancy while preventing loops through intelligent parameter-driven path selection.
2Object-affected harmful factors
If conventional spanning tree protocols are used to avoid loops, then loop formation is prevented, but the protocols fail to adequately address network overload and traffic optimization
Solution Approach 1:
The patent introduces dynamic path selection based on cost metrics that can change according to network conditions. Rather than static spanning trees, the system dynamically adjusts which paths are used based on real-time cost information, enabling both loop prevention and traffic optimization. This dynamic approach allows the network to adapt to changing conditions while maintaining both goals.
Solution Approach 2:
The patent adds a new dimension to conventional spanning tree protocols by introducing cost-based shortest-path calculations. This additional dimension of cost-aware path selection enables the system to optimize traffic not only for loop prevention but also for load balancing and traffic engineering, going beyond the capabilities of conventional protocols.
3Productivity
If control information is embedded within packet headers, then forwarding table population efficiency is improved, but packet header structure becomes more complex
Solution Approach 1:
The patent embeds control information within existing packet header fields, effectively nesting additional functionality within the existing structure. By utilizing available header space for control information, the system achieves efficient forwarding table population without requiring completely new header formats, thus limiting the increase in structural complexity.
Solution Approach 2:
The patent makes packet headers multi-functional by using them to carry both traditional routing information and control information for spanning tree protocols. This universal use of header fields allows the system to achieve multiple goals (forwarding efficiency and loop prevention) without proportionally increasing header complexity.
Data Source
AI summary
In accordance with on example, a control packet from a first network device is received at a second network device. The control packet comprises, embedded in a source address field of the control packet, control information for a spanning tree associated with a root network device. The control information is extracted from the source address field of the control packet in order to populate a shortest-path forwarding table. A data packet from the first network is then received by the second network device. Based on the shortest-path forwarding table and a media access control (MAC) address table, a host device to which the data packet is directed is identified. Finally, based on the information in the shortest-path forwarding table and the MAC address table, the data packet is forwarded to a network device directed connected to the host device.