Decision Service Pre-execution for Timeout Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Decision services in production computing environments frequently fail or timeout due to processing node timeouts, API failures, and unresponsive calls, leading to system errors, user losses, and inefficient use of computing resources.
Innovation Solution
Implementing intelligent pre-execution of decision service strategies by parsing and analyzing data processing pathways to determine available data, caching initial decisions, and executing them before receiving the actual data request, thereby reducing the need for real-time data loading and enhancing system availability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If all required data is loaded in real-time for decision service execution, then decision accuracy is improved, but system timeout and failure rates increase
Solution Approach 1:
The system performs preliminary data loading and strategy parsing before the decision service is actually invoked. Available data is pre-loaded into memory and strategy pathways are pre-analyzed during idle periods or upstream notification events, so that when the decision service needs to execute, the data is already prepared and ready, eliminating real-time loading delays that cause timeouts.
Solution Approach 2:
The strategy execution is divided into multiple independent pathways, each with its own data requirements. The system identifies and executes only the necessary pathways based on available data, rather than loading and processing all possible data. This segmentation allows the system to make decisions with partial data when appropriate, reducing loading time and increasing availability.
2Measurement precision
If comprehensive data loading is performed for every decision request, then decision quality is improved, but processing time increases
Solution Approach 1:
The system loads and processes only the portion of data necessary for the current decision context, rather than comprehensively loading all available data. By analyzing strategy pathways and identifying minimal required data sets, the system achieves sufficient decision quality with reduced processing time, avoiding the overhead of loading unnecessary data.
Solution Approach 2:
Data that is likely to be needed for future decisions is pre-loaded into memory during idle periods or triggered by upstream notifications before actual decision requests arrive. This preliminary preparation reduces the processing time for subsequent decisions by eliminating real-time data loading operations.
3Speed
If real-time data processing is performed for all requests, then responsiveness is improved, but resource utilization efficiency deteriorates
Solution Approach 1:
The system uses upstream notification events and idle processing periods to automatically prepare data and analyze strategies without waiting for actual decision requests. This self-service approach allows the system to perform useful work during otherwise wasted time, improving resource utilization without compromising responsiveness when decisions are actually needed.
Solution Approach 2:
Instead of continuously processing all data for every request, the system performs data loading and strategy analysis periodically during idle periods or in response to upstream notifications. This periodic action pattern allows the system to maintain responsiveness for actual decisions while significantly reducing overall computing resource consumption by avoiding continuous unnecessary processing.
Data Source
AI summary
There are provided systems and methods for intelligent pre-execution of decision service strategies for availability during data requests. A service provider, such as an electronic transaction processor for digital transactions, may utilize different decision services that implement rules and artificial intelligence models for decision-making of data including data in production computing environment. A decision service may normally be invoked to execute a strategy for processing a request when that request is received or arrives at the decision service. To provide faster processing and higher availability of such services, an intimation call may be received at the decision service from an upstream service, which initiates a process to pre-execute the strategy based on available data at the time of the intimation call. Pathways of rules capable of being completed may be executed using the available data and the result may be stored for later merging with further strategy execution.