Automated IT Capacity Forecasting via ARIMA Time Series Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current capacity planning methods for IT systems require manual intervention and statistical knowledge, making them inefficient for automated hardware upgrading and prone to errors in predicting future resource needs, especially in dynamic IT environments.
Innovation Solution
An automated time series prediction method based on the Box and Jenkins algorithm, enhanced with pre-processing stages to identify anomalies and seasonal components, and ARMA modeling, which can accurately forecast hardware needs without requiring manual intervention or deep statistical knowledge.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual intervention and statistical knowledge are used in capacity planning, then prediction accuracy can be improved through expert judgment, but the complexity and time consumption of the planning process increases
Solution Approach 1:
The system enables automated capacity planning where the algorithm independently performs data collection, analysis, and prediction without requiring manual statistical expertise. The Box-Jenkins ARIMA algorithm automatically identifies model parameters and generates forecasts, allowing the system to serve itself rather than relying on external expert intervention.
Solution Approach 2:
The patent replaces manual statistical analysis (mechanical human process) with an automated computational algorithm. The Box-Jenkins ARIMA model substitutes human experts' judgment with a systematic mathematical approach that automatically processes time series data to generate capacity predictions, eliminating the need for manual statistical knowledge.
2Extent of automation
If traditional time series analysis methods are used, then historical data can be analyzed to predict future needs, but the automation level remains low requiring human intervention
Solution Approach 1:
The system performs preliminary actions by pre-processing historical data and pre-configuring the ARIMA model parameters before actual prediction is needed. The algorithm automatically prepares the time series data, identifies seasonal patterns, and sets up the forecasting model in advance, so that when prediction is required, the system can generate results quickly without extensive real-time computation.
3Productivity
If automated prediction methods are implemented, then the upgrading process can be fully automated, but the precision of predicting hardware needs may deteriorate without expert tuning
Solution Approach 1:
The patent implements parameter changes by automatically adjusting the ARIMA model parameters (p, d, q values) based on the characteristics of the input time series data. The system dynamically modifies these parameters to optimize prediction accuracy for different hardware metrics and time patterns, enabling automated tuning that maintains precision while achieving full automation of the upgrading process.
Data Source
AI summary
Embodiments provide a method for performing an automatic execution of a Box and Jenkins method for forecasting the behavior of said dataset. The method may include pre-processing the dataset including providing one or more missing values to the dataset, removing level discontinuities and outliers, and removing one or more last samples from the dataset, obtaining a trend of the pre-processed dataset including identifying and filtering the trend out of the dataset based on a coefficient of determination methodology, detecting seasonality to obtain a resulting stationary series including computing an auto correlation function of the dataset, repeating the detecting step on an aggregate series of a previous dataset, and removing detected seasonality based on a seasonal differencing process, and modeling the resulting stationary series under an autoregressive-moving-average (ARMA) model.