Intelligent Server Routing with Dynamic Ranking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional routing systems for e-commerce servers are 'dumb' and fail to reroute requests from non-performant servers, rely on historic biases for ranking, do not account for server dependencies, and lack fraud detection, leading to suboptimal performance and security issues.
Innovation Solution
An intelligent server router system that dynamically ranks servers based on performance and response information, re-routes requests in real-time, examines server dependencies, and detects fraudulent behavior using automated test requests.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional routing systems are used to route requests between redundant servers, then the system structure is simple and easy to implement, but the routing is 'dumb' and cannot re-route requests from non-performant servers, leading to suboptimal performance
Solution Approach 1:
The patent implements feedback mechanisms where the routing system monitors server performance metrics (response time, error rates, throughput) and uses this information to dynamically adjust request routing. The system continuously collects performance data from servers and updates routing decisions based on real-time feedback, enabling intelligent re-routing from non-performant servers while maintaining reliable request processing.
Solution Approach 2:
The routing system transitions from static to dynamic operation by continuously adjusting routing decisions based on real-time server performance. The system dynamically modifies request distribution across servers according to changing performance conditions, allowing optimal load balancing and failover without requiring complex manual intervention, thus improving reliability while managing complexity through automation.
2Stability of the object's composition
If servers are ranked based on historic bias with large sample sizes, then the ranking is stable and reliable, but the ranking is slow to adapt to unexpected server failures or performance changes
Solution Approach 1:
The ranking system dynamically adjusts server rankings based on real-time performance metrics rather than relying solely on historic data. The system incorporates weighting factors that can emphasize recent performance more heavily, allowing the ranking to adapt quickly to unexpected server failures or performance changes while maintaining stability through continuous monitoring and iterative updates of server performance profiles.
Solution Approach 2:
The system changes the parameters used for ranking by incorporating multiple dynamic metrics (response time, error rates, throughput) with adjustable weights. This allows the ranking to balance stability from historic data with adaptability to current performance conditions, enabling the system to respond appropriately to both gradual performance trends and sudden server failures.
3Reliability
If traditional routing systems only monitor redundant servers, then the monitoring scope is limited and simple, but the system cannot detect failures in dependent servers that would cause redundant servers to fail
Solution Approach 1:
The monitoring system performs preliminary monitoring of dependent servers before they are needed by redundant servers. By proactively collecting performance data from dependent servers and establishing their operational status in advance, the system can detect potential failures before they affect redundant servers, allowing for preventive routing adjustments and maintaining reliability without requiring complex real-time monitoring of all dependencies.
Solution Approach 2:
The system introduces an intermediary monitoring layer that collects performance information from both redundant servers and their dependent servers. This intermediary monitoring mechanism aggregates data from the entire server hierarchy and presents it to the routing system, enabling comprehensive reliability monitoring while managing complexity through centralized data collection and processing rather than direct complex interconnections between all servers.
4Reliability
If traditional routing systems do not examine response substance, then the routing is fast and simple, but the system cannot detect servers returning error messages or fraudulent responses
Solution Approach 1:
The system applies partial examination of response substance by focusing monitoring on specific critical indicators (error codes, response status, content validity) rather than analyzing every aspect of server responses in detail. This selective monitoring approach enables detection of fraudulent responses and server errors without requiring comprehensive analysis of all response data, thus maintaining fast routing while improving reliability through targeted validation of response quality.
Data Source
AI summary
The present application relates to methods and systems for intelligently routing requests to one of a plurality of redundant servers. The methods and systems route the requests to a most highly ranked redundant server. The redundant servers are dynamically rank according to ranking information, wherein the ranking information may include server performance information, server response information, next step performance information, next step response information, historic information and other like information. Next step response information and next step performance information corresponds to information pertaining to servers the plurality of redundant servers depend upon. The methods and systems may further randomly re-rank the plurality of redundant servers. The methods and systems may further re-route unprocessed requests meeting non-performance criteria.