Machine-Learned Models for ERP Database Growth Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Enterprise Resource Planning (ERP) software databases face challenges in predicting growth patterns and managing database size effectively, leading to potential performance issues and increased storage costs due to the inability to accurately forecast future technological needs and table size.
Innovation Solution
The implementation of machine learning models, specifically a time field classification model and a growth driver classifier, to predict growth patterns in ERP software databases by analyzing time and category fields, allowing for precise forecasting of database size and recommending archiving operations to prevent performance degradation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional database management methods are used without machine learning prediction, then database operations are simple, but database growth cannot be accurately predicted leading to performance issues and increased storage costs
Solution Approach 1:
The system performs preliminary actions by training machine learning models on historical database growth data before actual prediction is needed. The models are pre-trained to recognize growth patterns, enabling accurate predictions when deployed without requiring complex real-time analysis during database operations.
Solution Approach 2:
Machine learning models serve as intermediaries between raw database data and actionable insights. The models translate complex database growth patterns into predictable trends, acting as a mediator that simplifies the relationship between historical data and future predictions while maintaining high accuracy.
2Measurement precision
If machine learning models are implemented to predict database growth, then prediction accuracy improves, but computational resources and processing time increase
Solution Approach 1:
The system performs computationally intensive model training in advance during off-peak hours or when resources are available, storing the trained models for later use. This preliminary action transfers computational burden from runtime to training time, reducing real-time resource consumption while maintaining prediction accuracy.
Solution Approach 2:
The system implements a two-stage prediction approach where a lightweight model provides quick predictions and a more comprehensive model provides detailed analysis only when needed. This partial action strategy balances computational resources by applying full analytical power selectively rather than continuously.
3Productivity
If database size is allowed to grow without prediction and management, then storage capacity is maximized in the short term, but database performance degrades and storage costs increase
Solution Approach 1:
The system performs preliminary archiving actions based on predicted growth patterns before performance degradation occurs. By proactively identifying tables and records that will soon exceed optimal size thresholds, the system archives data in advance, maintaining database performance without reactive emergency interventions.
Solution Approach 2:
The system implements continuous feedback loops where prediction results inform archiving decisions, which in turn affect actual database growth patterns. This feedback mechanism allows the system to adjust archiving strategies based on actual versus predicted growth, optimizing the balance between database size and performance over time.
Data Source
AI summary
In an example embodiment, machine learning models are trained and used to predict a growth classification of time fields and category fields of application tables of Enterprise Resource Planning (ERP) software databases. These predictions can then be used to forecast future technological needs or the future table size more precisely.