Error Handling Tool for Synchronous Request Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional error handling in online synchronous calls leads to unnecessary reissuing of entire requests when component requests fail, causing inefficiencies as the number of computers and mobile devices increases, resulting in servers and resources redoing previously completed work.
Innovation Solution
An error handling tool that uses machine learning to detect and address failures by updating reprocessing rules, allowing component requests to be reissued before the user request times out, and adapts timeouts and failure conditions to reduce future failures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If the entire request is reissued when a component request fails, then the failure is handled simply, but system resources are wasted by redoing previously completed work
Solution Approach 1:
The patent segments the request into multiple component requests, each with independent timeout handling. When a component request fails, only that specific component is reissued with adjusted parameters, while other successful component requests are retained and assembled into the final response. This segmentation prevents the need to reissue the entire request, thereby avoiding waste of system resources on previously completed work.
2Loss of energy
If component requests are reissued individually, then resource waste is reduced, but the complexity of error handling increases
Solution Approach 1:
The patent applies parameter changes by adjusting timeout values and reconnect parameters dynamically based on failure conditions. When a component request fails, the system modifies the timeout parameter for subsequent reissues of that specific component request. This parameter adjustment mechanism, guided by machine learning models, manages the complexity of individual component reissuing by automating parameter optimization.
Solution Approach 2:
The patent implements feedback mechanisms where the system learns from previous failures and adapts future timeout and failure condition settings. Machine learning models analyze historical failure data to optimize error handling strategies, reducing the manual complexity of managing multiple component requests with different timeout parameters.
3Reliability
If timeouts are extended for component requests, then failure rate decreases, but the total response time increases
Solution Approach 1:
The patent employs dynamic timeout adjustment rather than using fixed timeouts. When a component request fails, the system dynamically extends the timeout parameter for subsequent reissues of that specific component. This dynamic adjustment allows the system to balance reliability and response time by extending timeouts only when necessary, based on the specific failure conditions and component behavior patterns.
Solution Approach 2:
The system changes timeout parameters adaptively based on failure analysis. Machine learning models analyze the context of each failure and adjust the timeout parameter accordingly, extending it only when the failure pattern suggests a temporary or recoverable issue. This parameter change strategy improves reliability without unnecessarily inflating response times for all requests.
Data Source
AI summary
An apparatus includes a memory and a hardware processor. The memory stores a plurality of reprocessing rules. The processor receives a request message from a user device. The processor communicates a second request to a first resource and a third request to a second resource. The processor determines that a response to the second request was not received. The processor increases the first timeout. The processor communicates the second request to the first resource after increasing the first timeout, receives a response to the second request, and determines that a response to the third request was not received. The processor increases the reconnect parameter. The processor communicates the third request to the second resource after increasing the reconnect parameter, receives a response to the third request, generates a response message to the request message, and communicates the response message.