Rate Limiter Manager for Network Flow Aggregation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing rate limiting systems in network communications face inefficiencies, with hardware rate limiters being limited in capacity and software rate limiters consuming excessive computational resources, leading to congestion and power consumption issues in data centers.
Innovation Solution
A system and method that employs a rate limiter manager module to aggregate flows based on aggregation attributes and assign them to rate limiters, allowing for efficient distribution of flows across hardware and software rate limiters, optimizing resource usage and reducing congestion by forming aggregate flows and adjusting rate limits dynamically.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If hardware rate limiters are used, then processing speed is improved, but the number of flows they can govern is limited
Solution Approach 1:
The patent segments flows into different categories based on their characteristics (e.g., aggregate flows vs. individual flows). Hardware rate limiters are assigned to govern aggregate flows while software rate limiters handle individual flows, allowing the system to scale to many more flows while maintaining high-speed hardware enforcement where applicable.
Solution Approach 2:
The patent introduces a rate limiter manager as an intermediary component that sits between the hardware rate limiters and software rate limiters. This manager dynamically assigns flows to appropriate rate limiters, coordinating their work to maximize both processing speed and flow governance capacity.
2Adaptability or versatility
If software rate limiters are used, then the number of flows they can govern is increased, but computational resources are excessively consumed
Solution Approach 1:
The patent merges multiple individual flows into aggregate flows based on common characteristics (e.g., same destination, same traffic type). By governing these aggregate flows together using hardware rate limiters or shared software rate limiters, the system reduces the total number of rate limiting operations needed, thereby reducing computational resource consumption while still governing a large number of individual flows.
Solution Approach 2:
The patent applies rate limiting at an aggregate level rather than enforcing individual rate limits on every single flow. This partial action approach (limiting the aggregate rather than each individual flow) significantly reduces computational overhead while providing sufficient rate control for most scenarios.
3Productivity
If flows are aggregated into aggregate flows, then resource utilization is optimized, but the complexity of flow management increases
Solution Approach 1:
The patent creates a universal rate limiter manager that handles multiple functions: flow classification, aggregation decisions, rate limiter selection, and dynamic reassignment. This single multi-functional component manages the complexity centrally, allowing individual rate limiters to remain simple while the system as a whole achieves optimized resource utilization through intelligent aggregation.
Data Source
AI summary
A system includes a plurality of rate limiters and a rate limiter management module. The rate limiter management module causes a processor to analyze and compare a plurality of flows each having at least one aggregation attribute, to aggregate a first group of the plurality of flows into a first aggregate flow based on the at least one aggregation attribute, and to assign the first aggregate flow to a rate limiter having at least one membership criterion. The first aggregate flow includes flows having the at least one aggregation attribute that satisfies the at least one membership criterion.