Data Transformer Energy Prediction via Rule Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data transformation methods consume significant system resources and face inaccuracies in predicting energy consumption due to reliance on historical load predictions, leading to inefficient energy management.
Innovation Solution
A data transformation method that imports data transform rules, predicts resource energy consumption parameters based on source and destination data definitions, and deploys an energy optimization policy for the data transform node server, analyzing energy consumption in three phases to dynamically adjust resource usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If dynamic voltage and frequency scaling (DVFS) technique is used to predict future load conditions according to history load, then energy consumption can be reduced, but prediction accuracy becomes very inaccurate due to large deviation between historical load and actual load
Solution Approach 1:
The patent performs preliminary analysis of data transform rules, source data definitions, and destination data definitions before actual data transformation occurs. By predicting energy consumption parameters in advance based on these definitions rather than historical load, the system can proactively optimize resource allocation and deploy appropriate energy consumption optimization policies, avoiding the inaccuracies of historical load prediction while still achieving energy reduction goals
Solution Approach 2:
The patent replaces the mechanical approach of predicting future load based on historical patterns with a systematic analysis approach that evaluates actual data transform requirements. Instead of using past performance data to infer future needs, the system directly analyzes the characteristics of the upcoming data transformation task (rules, source definitions, destination definitions) to predict energy consumption, substituting the predictive mechanism with a more accurate analytical one
2Adaptability or versatility
If data transform is performed to transform data from one representative form to another, then data integration and software product integration can be realized, but system energy consumption increases significantly
Solution Approach 1:
The patent performs preliminary energy consumption prediction by analyzing data transform rules, source data definitions, and destination data definitions before the actual data transformation process. This allows the system to determine optimal resource allocation and deploy energy consumption optimization policies in advance, ensuring that data integration tasks can be completed with minimized energy consumption while maintaining full adaptability across different data formats and systems
Solution Approach 2:
The patent dynamically adjusts resource allocation parameters (such as computing resources, memory, and processing power) based on predicted energy consumption parameters. By changing these parameters according to the specific requirements of each data transformation task and the predicted energy consumption, the system achieves optimal balance between data integration capability and energy consumption, allowing versatile data transformation while reducing overall energy usage
Data Source
AI summary
A data transform method and a data transformer. The method includes: importing a data transform rule; acquiring from the data transform rule a source data definition, a destination data definition and a data transform rule definition; predicting resource energy consumption parameters of a data transform node server according to the source data definition, the destination data definition and the data transform rule definition; and deploying a resource energy consumption optimization policy of the data transform node server according to the predicted resource energy consumption parameters of the data transform node server.