Communication Performance Analysis Using Multidimensional Scaling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for analyzing communication performance in computer systems are inefficient, making it difficult to intuitively understand communication time data, especially when dealing with a large number of computers, and fail to automatically detect potential issues or prioritize computers for trouble identification.
Innovation Solution
A communication performance analyzing program that uses statistical analysis techniques such as metrical and non-metrical multidimensional scaling, Sammon mapping, clustering, principal component analysis, and independent component analysis to acquire and analyze communication time data, allowing for the grouping of computers and observation items, and displaying results in a way that facilitates understanding of performance trends and issue detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If visual inspection of communication time data is performed manually, then the analysis process is simple, but it becomes difficult to intuitively understand communication performance when the number of computers is large
Solution Approach 1:
The patent transforms high-dimensional communication time data (N×(N-1)/2 combinations) into a low-dimensional visual representation using multidimensional scaling techniques. This allows the complex communication performance data to be displayed in a 2D or 3D space where spatial relationships reflect communication efficiency, enabling intuitive understanding without manual inspection of all data points.
Solution Approach 2:
The patent creates a virtual copy or model of the communication network topology where nodes represent computers and edge lengths represent communication times. This visual model allows observers to intuitively grasp communication performance patterns without directly examining the raw data, effectively copying the essential characteristics into a more comprehensible form.
2Reliability
If all communication time data for N computers is collected, then complete communication performance information is obtained, but the data volume and complexity increase significantly
Solution Approach 1:
The patent extracts only the essential characteristics of communication performance from the complete data set. By using multidimensional scaling and clustering algorithms, it identifies and highlights the most significant patterns and outliers, extracting key performance indicators without needing to process and display all raw data points, thus reducing complexity while maintaining reliability.
Solution Approach 2:
The patent combines multiple communication time measurements into aggregated visual representations. By merging data from multiple observation items and communication modes into a unified visual model, it reduces the overall data complexity while preserving the essential performance characteristics through statistical aggregation and pattern recognition.
3Reliability
If multiple observation items (communication modes and data lengths) are analyzed, then comprehensive performance evaluation is achieved, but the number of data sets increases to N1×N2
Solution Approach 1:
The patent creates a universal visual framework that can simultaneously represent multiple observation items and communication modes within a single multidimensional scaling model. This multi-functional approach allows the same visual representation technique to handle diverse data types (different communication modes, data lengths, and time periods) uniformly, reducing the need for separate analysis procedures for each data set.
Data Source
AI summary
A communication performance analyzing program, a communication performance analyzing apparatus and a communication performance analyzing method make it possible to highly reliably grasp the communication performance of a computer system by automatically analyzing the communication performance. The communication performance analyzing program that causes a computer to analyze a data tendency of communication performance of a plurality of execution periods of a computer system formed by connecting a plurality of computers by a network, the program comprises a communication time acquisition step S2 that acquires communication time data among the computers of the computer system and a statistical analysis step S3 that analyzes the tendency of communication performance data of each execution period of the computer system, using the communication time among the computers, by statistically analyzing the communication time data acquired by the communication time acquisition step.