API Platform Dynamic Backend Scaling for Load Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data linkage systems face challenges in efficiently managing data processing capacity based on the load of API requests, particularly in determining the appropriate processing resources needed for varying predicted processing times and response data amounts, which can lead to inefficiencies and potential system failures.
Innovation Solution
The proposed data linkage system incorporates an API platform that uses a queue and backend classification system based on predicted processing time and response data amount levels, dynamically adjusting processing capacity by increasing or reducing backend instances according to specific thresholds, and includes mechanisms to handle failures by stopping processing in case of data storage system maintenance or failure.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the API platform processes all API requests with fixed capacity, then system simplicity is maintained, but processing efficiency deteriorates under varying load conditions
Solution Approach 1:
The API platform dynamically adjusts processing capacity by classifying backend instances into different types (first type for normal processing, second type for high-priority or resource-intensive processing) and selectively activating appropriate backend types based on request characteristics and system load conditions, enabling flexible adaptation to varying demands
2Speed
If the system increases processing capacity for all requests, then processing speed improves, but resource waste increases during low-load periods
Solution Approach 1:
The system changes operational parameters by adjusting the number and type of active backend instances based on load conditions, using classification rules to determine whether to activate first-type or second-type backends, thereby optimizing processing speed while minimizing resource consumption during low-demand periods
3Adaptability or versatility
If the API platform uses a single processing mode, then ease of operation is maintained, but adaptability to different data volumes and processing times deteriorates
Solution Approach 1:
The backend processing capacity is segmented into multiple types (first type and second type) with different characteristics, allowing the system to select appropriate processing modes based on request attributes such as data volume, processing time requirements, and priority levels, thereby achieving adaptability while maintaining automated operation
4Productivity
If the system continuously monitors and adjusts processing capacity, then responsiveness to load changes improves, but system complexity and overhead increase
Solution Approach 1:
The API platform implements self-service through automated classification rules that evaluate incoming requests and automatically determine the appropriate backend type to activate, eliminating the need for manual intervention or complex external control mechanisms while maintaining high responsiveness to load changes
Data Source
AI summary
A data linkage system and an API platform are capable of behaving according to a load of processing executed in response to a request for provision of an API. The data linkage system includes the API platform that provides the API for acquiring data, which is based on data collected by a data collection system and stored in a data storage system for storing data held by an information system, from the data storage system. The API platform determining whether the number of the data that has not been subjected to processing to acquire the data from the data storage system in response to a request for provision of the API satisfies a specific condition, and changing capacity of the processing according to a result of the determination.