Traffic Manager Routing Network Traffic Without Centralized Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing traffic management systems face challenges in maintaining consistent routing of network traffic when a traffic manager fails, leading to potential loss of connectivity due to the loss of information about which server to route requests to, especially in high-load scenarios with multiple traffic managers sharing the same IP address.
Innovation Solution
A traffic manager system that includes memory, a network interface, and a processor to receive and analyze requests from client computers, identify the responsible server based on request attributes, and route the requests independently without centralized or synchronized information, ensuring even distribution of load and responsibility across multiple traffic managers and servers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If multiple traffic managers share the same IP address to handle high load, then the system can serve more clients simultaneously, but when one traffic manager fails, the routing information is lost and connectivity is disrupted
Solution Approach 1:
The patent segments the routing responsibility by assigning each traffic manager a unique identifier (such as a MAC address or internal ID) that is embedded in the routing decisions. This segmentation allows the system to distinguish between different traffic managers even though they share the same IP address, enabling failover without loss of routing information.
Solution Approach 2:
The patent introduces an intermediary mechanism where traffic managers exchange routing information through a shared memory space or communication channel. When a traffic manager fails, the intermediary mechanism allows other traffic managers to retrieve updated routing information, maintaining connectivity without requiring centralized control.
2Reliability
If centralized routing control is used to maintain consistent routing information, then routing reliability is improved, but system complexity and single point of failure risk increase
Solution Approach 1:
The patent enables traffic managers to make autonomous routing decisions based on their own unique identifiers and the current state of the system. Each traffic manager independently determines routing decisions without requiring constant centralized control, reducing system complexity while maintaining reliability through distributed intelligence.
Solution Approach 2:
The patent changes the routing decision parameters from centralized control to distributed parameters embedded in each traffic manager's identity and configuration. This parameter transformation allows traffic managers to operate independently while maintaining consistent routing behavior, eliminating the need for complex centralized control systems.
3Productivity
If traffic managers operate independently without synchronization, then system performance and availability are improved, but routing consistency across failures may be compromised
Solution Approach 1:
The patent implements preliminary action by pre-configuring each traffic manager with a unique identifier and routing capabilities before failure occurs. This preliminary setup enables traffic managers to operate independently and immediately take over routing functions when failures happen, without requiring real-time synchronization or negotiation during critical moments.
Data Source
AI summary
A traffic manager (121-124) for, and a method of, routing network traffic to a plurality of server computers (131-138). The traffic manager includes a network interface (204) and a processor (201). The processor is configured to receive network traffic comprising a request (701, 702) from a client computer (102-105) via the network interface. The processor is configured to then identify, based on attributes of the request, a server computer that is responsible for servicing the request (a responsible server). The processor is configured to then route the request to the responsible server using the network interface.