Load Balancing Virtual Point-to-Point Links Across Provider Edge Nodes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current load-balancing processes do not effectively distribute traffic across multiple intermediate network nodes, leading to congestion on some links while others remain underutilized, particularly in provider edge networks.
Innovation Solution
Implementing a method for balancing data packet traffic loads over virtual point-to-point links by bundling and managing virtual links across multiple provider edge nodes using a load balancing process that associates data-link addresses with physical and virtual links, and selecting optimal virtual links for packet forwarding based on load balancing considerations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If current load-balancing processes are used, then traffic distribution is simple, but traffic congestion occurs on some links while others remain underutilized
Solution Approach 1:
The patent changes the parameters used for load balancing from simple hash-based destination address mapping to a more sophisticated scheme that considers multiple factors including data-link addresses, virtual link identifiers, and real-time link status. This allows better traffic distribution across bundled links while maintaining compatibility with existing network protocols.
Solution Approach 2:
The invention implements a feedback mechanism where the load-balancing process continuously monitors link utilization and dynamically adjusts traffic distribution. The system tracks which virtual links are currently in use, their capacity, and their status, then uses this information to make intelligent forwarding decisions, preventing both congestion and underutilization.
2Reliability
If multiple virtual links are bundled across multiple provider edge nodes, then network redundancy is improved, but traffic distribution across these links becomes uneven
Solution Approach 1:
The patent segments traffic into different virtual links based on multiple criteria including data-link address, virtual link identifier, and link status. By dividing the traffic flow appropriately across segmented virtual links, the system achieves both redundancy and efficient utilization. Each virtual link serves a specific purpose while collectively providing robust traffic distribution.
Solution Approach 2:
The load-balancing process dynamically adapts to changing network conditions by monitoring virtual link status in real-time. When links become congested or fail, the system dynamically reroutes traffic through alternative virtual links, maintaining both redundancy and efficiency. This dynamic behavior ensures optimal traffic distribution regardless of network state.
3Ease of operation
If simple destination address hashing is used for load balancing, then implementation is easy, but traffic concentration occurs on specific links
Solution Approach 1:
The patent creates a universal load-balancing mechanism that serves multiple functions: it maintains backward compatibility with simple hash-based approaches while simultaneously providing sophisticated traffic distribution based on multiple parameters. The system can operate in different modes depending on network requirements, combining the ease of simple hashing with the effectiveness of multi-parameter decision-making.
Data Source
Figure 1A
Figure 1B
Figure 2
AI summary
Techniques for load balancing network traffic over virtual point-to-point (PPP) data links at a data link layer include determining a particular group of virtual PPP data links that reach a particular local area network of one or more network nodes. A particular data link address for a particular node of the particular local area network is associated with each virtual PPP data link of the particular group. When a data packet is received that includes data that indicates the particular data link address as a destination; one virtual PPP data link of the group is determined to balance traffic among the group. The data packet is sent only over the one virtual PPP data link. These techniques enable virtual PPP data links through two or more intermediate network nodes, such as a pair of provider edge nodes, to be bundled for load balancing purposes.