Cluster Drawing on Divided Display Region for Correlation Diagrams
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In large-scale systems, the distribution of clusters in correlation diagrams results in extremely large clusters with many metrics and numerous small clusters with few metrics, leading to poor visibility due to significant size differences, making it difficult to overview the correlation diagram effectively.
Innovation Solution
A system analysis device and method that divide a display region into areas where each area's size is equal to or larger than the next, allocating clusters in decreasing order of metrics, ensuring the number of clusters increases with area size, allowing for improved visibility by maintaining larger areas for larger metric clusters and smaller areas for smaller metric clusters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If area allocation is proportional to the number of metrics in each cluster, then large clusters receive adequate display space, but small clusters become too small to be visible
Solution Approach 1:
The patent segments the display region into multiple divided regions (first divided region, second divided region, etc.) with different area sizes. Each divided region is assigned to accommodate clusters of different metric counts, thereby segmenting the display task by size categories rather than using a uniform proportional allocation.
Solution Approach 2:
The patent applies local quality by allocating different area sizes to different divided regions based on the characteristics of clusters they contain. The first divided region (larger area) is used for clusters with more metrics, while the second divided region (smaller area) is used for clusters with fewer metrics, optimizing visibility for each local context.
2Loss of information
If proportional allocation is used, then the correlation diagram reflects the actual distribution of metrics, but the diagram becomes difficult to overview due to extreme size differences
Solution Approach 1:
The patent segments clusters into different groups based on their metric counts and assigns them to different divided regions. This segmentation prevents extreme size differences from dominating the entire diagram, making it easier to overview while still representing the distribution characteristics through the structured arrangement of divided regions.
3Loss of information
If small clusters are given adequate space, then all clusters become visible, but large clusters lose their proportional representation
Solution Approach 1:
The patent introduces a new dimension of organization by dividing the display region into multiple divided regions with different area sizes. Instead of competing for space in a single uniform dimension, clusters are arranged across different divided regions, allowing small clusters to be visible in the second divided region while large clusters maintain their presence in the first divided region.
Data Source
AI summary
Clusters of metrics in a system are stored. A display region is divided into n divided regions in such a way that an area of a divided region i (1≤i≤n) is equal to or larger than an area of a divided region i+1. Each cluster is allocated to the divided region i sequentially selected from i=1, in the decreasing order of the number of metrics contained in each of the clusters, in such a way that the number of the clusters allocated to the divided region i+1 is equal to or more than the number of clusters allocated to the divided region i. The cluster allocated to the divided region i is drawn in the divided region i.