Session Traffic Congestion Control via Dynamic Redundancy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
As session-based network traffic increases, existing technologies face challenges in providing effective redundancy and load balancing among clusters of session handling servers, particularly in ensuring continuous operation and efficient data processing across multiple servers.
Innovation Solution
Implementing a congestion control method that utilizes a router or cluster manager to monitor server utilization, schedule data traffic, and dynamically adjust redundancy factors, ensuring that each server in the cluster handles corresponding data loads while maintaining redundancy for fault tolerance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If redundancy capacity is increased to ensure continuous operation during server failures, then system reliability is improved, but available processing capacity for data traffic is reduced
Solution Approach 1:
The patent implements dynamic adjustment of redundancy factors based on real-time server utilization monitoring. The system continuously adapts the redundancy level by adjusting the ratio of unused capacity reserved for failover, allowing the cluster to optimize between reliability and productivity based on current operational conditions rather than maintaining fixed redundancy levels
Solution Approach 2:
The system changes the parameter of redundancy factor (ri) for each server based on monitored utilization patterns. By modifying these parameters dynamically, the system can increase redundancy when needed for reliability while maintaining higher productivity when servers are underutilized, thus resolving the contradiction between these two opposing requirements
2Productivity
If load balancing is implemented to distribute traffic evenly across servers, then system efficiency is improved, but complexity of traffic scheduling increases
Solution Approach 1:
The patent employs feedback mechanisms where the cluster manager continuously monitors server utilization metrics and uses this information to dynamically adjust traffic scheduling decisions. This feedback-driven approach enables efficient load balancing by routing traffic to appropriately utilized servers while keeping the scheduling logic relatively simple and adaptive rather than complex and predetermined
3Productivity
If server utilization is increased to maximize processing capacity, then productivity is improved, but system reliability deteriorates due to reduced redundancy
Solution Approach 1:
The system dynamically adjusts the operational utilization level of each server based on the configured redundancy factors and real-time cluster conditions. When a server fails, the system automatically redistributes its load to remaining servers while maintaining appropriate redundancy levels, thus allowing high productivity during normal operation while preserving reliability through adaptive resource allocation
Data Source
AI summary
A method includes establishing an expected traffic load for a plurality of servers, wherein each server has a respective actual capacity. The method further includes limiting the actual capacity of each server to respective available capacities, wherein a combined available capacity that is based on the available capacities corresponds to the expected traffic load. The method also includes dynamically altering the respective available capacity of the servers based on the failure of at least one server.