WCMP Weight Distribution for Network Throughput Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Network collisions caused by multiple data packets requesting the same destination lead to congestion, higher latency, and lower throughput due to the lack of efficient load balancing across multiple paths with varying bandwidths in network switches.
Innovation Solution
Implementing Weighted Cost Multi-Path (WCMP) routing with a method to identify original WCMP groups, determine data throughput percentages, and create 'skinny' WCMP groups by proportionately reducing weights to minimize throughput degradation within a threshold, thereby optimizing resource usage in forwarding tables.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple data packets request the same destination through network switches with multiple paths, then routing flexibility is improved, but packet collisions increase leading to higher latency and lower throughput
Solution Approach 1:
The patent segments the load balancing task by creating separate WCMP groups for different destination prefixes, allowing independent weight distribution for each segment. This enables precise control over packet distribution to avoid collisions while maintaining routing flexibility.
Solution Approach 2:
The patent dynamically adjusts weight parameters in WCMP groups to optimize packet distribution. By changing weight values based on network conditions and destination prefixes, the system maintains high throughput while minimizing packet collisions through intelligent parameter tuning.
2Productivity
If WCMP groups are replicated in forwarding tables proportionally to assigned weights, then load balancing is improved, but hardware resources for storage increase
Solution Approach 1:
The patent divides the forwarding table into separate WCMP group entries for different destination prefixes, allowing the hardware to process each segment independently. This segmentation enables efficient load balancing while reducing the total number of entries that need to be stored and processed.
Solution Approach 2:
The patent applies partial replication by creating WCMP groups only for active destination prefixes rather than replicating all possible paths. This selective approach maintains effective load balancing for actual traffic patterns while significantly reducing hardware storage requirements.
3Quantity of substance
If weights are reduced to create skinny WCMP groups, then hardware resource consumption is reduced, but throughput degradation occurs
Solution Approach 1:
The patent optimizes weight parameters to create the minimum necessary WCMP group entries while maintaining acceptable throughput. By carefully tuning weight values, the system achieves reduced hardware resource consumption without significant throughput degradation.
Solution Approach 2:
The patent applies partial weight reduction by creating skinny WCMP groups only for less critical destination prefixes while maintaining full weight replication for critical paths. This selective approach reduces overall hardware resource consumption while preserving essential throughput performance.
Data Source
AI summary
The present technology considers network devices that include forwarding tables having a number of next-hop entries (e.g., egress ports) where it is possible that the egress port utilization can be load balanced using WCMP groups. In a WCMP group, its members are assigned weights representing an amount of data flow to distribute over a plurality of links for a given destination. This disclosure provides systems and methods to distribute weights assigned to members of WCMP groups while achieving a minimum target for network throughput.