Multidimensional Data Filtering in Star Schema Processing

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Multidimensional filtering of large data collections modeled in a star schema is resource-intensive, leading to long wait times due to the heavy use of resources required for processing.

Innovation Solution

A method and system that processes data by generating modified dimension and fact tables based on a data filter, where each column associated with a dimension table is filtered to produce a subset of data that satisfies the specified criteria, reducing resource usage and improving efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If multidimensional filtering is applied to large data collections in a star schema, then data accuracy and completeness are improved, but resource consumption increases and wait times lengthen

Engineering Contradiction:

Improvedata filtering accuracyVSAvoiddata processing speed

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The patent segments the filtering process into two distinct phases: (1) generating modified dimension tables by filtering dimension data according to filter criteria, and (2) using these modified dimension tables to filter fact table data. This segmentation allows dimension filtering to be performed once and reused across multiple fact table queries, improving overall processing efficiency while maintaining filtering accuracy.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary filtering of dimension tables before filtering the fact table. By pre-processing dimension data to create modified dimension tables that contain only relevant data according to filter criteria, the system reduces the complexity and resource requirements of subsequent fact table filtering operations.

Inventive Principle:

Principle #10Preliminary action

2Reliability

If multidimensional filtering processes large collections of data, then comprehensive results are produced, but resource usage increases heavily

Engineering Contradiction:

Improvefiltering completenessVSAvoidprocessing resource consumption

Core Design Contradiction:

ReliabilityVSUse of energy by moving object

Solution Approach 1:

The filtering operation is divided into dimension table filtering and fact table filtering stages. By filtering dimension tables first and storing the results as modified dimension tables, the system avoids repeatedly processing the same dimension data against multiple fact tables, thereby reducing overall resource consumption while maintaining filtering completeness.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent creates modified dimension tables as copies of the original dimension tables, filtered according to the specified criteria. These modified dimension tables serve as reusable reference data that can be applied to multiple fact table filtering operations without re-processing the original dimension data, reducing resource usage while ensuring consistent filtering results.

Inventive Principle:

Principle #26Copying

3Manufacturing precision

If traditional multidimensional filtering is used on star schema data, then all criteria are satisfied, but wait times become excessively long

Engineering Contradiction:

Improvefilter criteria satisfactionVSAvoidquery wait time

Core Design Contradiction:

Manufacturing precisionVSLoss of time

Solution Approach 1:

The system performs preliminary filtering of dimension tables to create modified dimension tables that encapsulate filter criteria results. This pre-processing step ensures that when fact tables are filtered, the system can efficiently join with pre-filtered dimension data rather than re-evaluating all dimension filter criteria for each fact table row, significantly reducing wait time while maintaining complete criteria satisfaction.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent separates dimension filtering from fact table filtering into distinct, sequential operations. By completing dimension table filtering first and storing results in modified dimension tables, the system eliminates redundant filter evaluations during fact table processing, reducing overall query wait time while ensuring all filter criteria are fully satisfied through the two-stage filtering process.

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS8145626B2Method and system for data processing using multidimensional filtering

Publication Date: 2012.03.27 SAP SE

AI summary

In one embodiment the present invention includes a method comprising receiving a data filter for filtering a collection of data, wherein the collection of data is configured as a star schema including a fact table and dimension tables. The data filter is applied against the dimension tables to generate a modified dimension table. The modified dimension tables are applied against the fact table to produce a modified fact table. The data filter is then applied against the modified fact table to generate a second modified fact table, which is the output of the process.