Time Sequence Pattern Estimator for Load Balancing

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods for extracting time sequence patterns from large-scale data lead to increased memory consumption and processing time due to uneven distribution of patterns across sub-groups, causing load imbalances during parallel distributed processing.

Innovation Solution

A data pattern analysis optimizer that estimates the upper limit of time sequence patterns, adjusts the number of items in each sub-group, and integrates sub-groups to balance processing load, using a grouping unit and estimator to optimize memory usage and support levels for efficient pattern extraction.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If time sequence data is divided into equivalent groups for parallel processing, then processing speed is improved, but load imbalance occurs causing certain groups to have significantly higher processing loads

Engineering Contradiction:

Improveprocessing speedVSAvoidload balance

Core Design Contradiction:

ProductivityVSEase of operation

Solution Approach 1:

The patent applies preliminary action by performing estimation of pattern numbers and predictive grouping before the actual parallel processing begins. The grouping unit divides time sequence data into sub-groups based on estimated pattern distribution, ensuring load balance is established in advance before processing starts, thereby avoiding load imbalance during execution

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent implements feedback through the estimator that calculates the number of time sequence patterns in each sub-group and provides this information back to the grouping unit. This feedback mechanism allows the system to adjust grouping strategies to achieve balanced load distribution across processing groups

Inventive Principle:

Principle #23Feedback

2Reliability

If the number of time sequence patterns is increased to extract more comprehensive patterns, then analysis completeness is improved, but memory consumption and processing time increase significantly

Engineering Contradiction:

Improveanalysis completenessVSAvoidprocessing time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent applies segmentation by dividing the time sequence data into multiple sub-groups with controlled pattern numbers. This allows comprehensive pattern extraction to be performed on smaller, manageable segments rather than attempting to process all patterns simultaneously in the entire dataset, thereby reducing processing time while maintaining completeness

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent changes parameters by adjusting the predefined number of items in time sequence patterns and the minimum support level based on estimated pattern distributions. This allows optimization of processing efficiency by setting appropriate parameter values that balance analysis completeness with processing time and memory constraints

Inventive Principle:

Principle #35Parameter changes

3Measurement precision

If the predefined number of items in time sequence patterns is increased to capture more complex patterns, then pattern extraction accuracy is improved, but the number of pattern types increases causing higher processing loads

Engineering Contradiction:

Improvepattern extraction accuracyVSAvoidprocessing complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent applies preliminary action by using the estimator to predict the number of pattern types that will be generated with different predefined item numbers before actual processing. This allows selection of appropriate parameter values in advance that achieve desired accuracy while controlling processing complexity

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent uses copying by creating multiple sub-groups that are processed in parallel, where each sub-group handles a portion of the pattern extraction task. This distributes the processing complexity across multiple simpler parallel tasks rather than one complex sequential task

Inventive Principle:

Principle #26Copying

Data Source

PatentUS11663210B2Data pattern analysis optimizer, and method of data pattern analysis optimization processing

Publication Date: 2023.05.30 TOSHIBA DIGITAL SOLUTIONS CORP

AI summary

An embodiment of a data pattern analysis optimizer includes a time sequence data memory, an estimator, a grouping unit, and a time sequence pattern extractor. The time sequence data memory stores a plurality of time sequence data made from items in time order. The estimator estimates the upper limit of the total number of types of time sequence patterns present in the time sequence data at a rate higher than a minimum support level, based on a respective rate of presence of each item, wherein each of the time sequence patterns present in the time sequence data is a predefined number of items. In case that the estimated upper limit exceeds an upper limit of the number of types of time sequence patterns as a maximum processing load to a computer, the grouping unit groups a plurality of time sequence data into sub-groups, based on a group of items having the increased number of items and gives the estimator instructions to perform estimation. The time sequence pattern extractor gives the computer instructs to extract the time sequence patterns for each of the sub-groups, in case that the estimated upper limit does not exceed the upper limit of the number of time sequence patterns.