Multi-OS Causal Analysis Program for Virtual Machine Delay Diagnosis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In systems with multiple virtual machines, analyzing the cause of processing delays is challenging due to the complexity of tracing operations across multiple operating systems, especially when the issue lies outside the virtual machine, such as in another virtual machine or the host OS, and existing methods are resource-intensive and require specialized knowledge.
Innovation Solution
A computer-readable recording medium stores an analysis program that acquires and totals sampling data from multiple operating systems at regular intervals, generates time-series data, and uses Bayesian estimation to analyze causal relationships between functions, allowing for the identification of the root cause of abnormalities without requiring extensive user intervention or recompilation of programs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If trace data is collected from multiple operating systems to analyze abnormality causes, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent segments the complex multi-OS system into independent sampling units, where each operating system independently samples its own function execution information. This segmentation allows parallel data collection without requiring complex coordination between OSs, thereby improving measurement precision while managing system complexity through modular independent operation.
Solution Approach 2:
The patent introduces a storage medium as an intermediary that receives and stores sampling results from multiple operating systems. This intermediary component decouples the complexity of multi-OS interaction from the analysis process, allowing each OS to write to the storage medium independently while the abnormality cause can be analyzed by examining the collected trace data without direct OS-to-OS communication overhead.
2Measurement precision
If sampling is performed at a short first time interval to capture detailed function execution, then measurement precision is improved, but processing overhead increases
Solution Approach 1:
The patent implements periodic sampling at a first time interval that is shorter than a predetermined threshold, creating a time-series dataset that captures function execution patterns without continuous monitoring. This periodic approach provides sufficient measurement precision to identify abnormality causes while avoiding the excessive processing overhead that would result from continuous or overly frequent sampling.
Solution Approach 2:
The patent changes the sampling parameter (time interval) dynamically by setting it to be shorter than a predetermined value based on the specific analysis needs. This parameter adjustment allows the system to achieve high measurement precision when necessary while maintaining lower processing overhead during normal operation, optimizing the balance between accuracy and resource consumption.
3Reliability
If comprehensive sampling data is collected from all operating systems, then reliability of analysis is improved, but loss of time increases due to data processing
Solution Approach 1:
The patent performs preliminary sampling and stores function execution information in a storage medium before abnormality analysis is required. By pre-collecting and organizing the trace data from multiple operating systems, the system ensures that when an abnormality occurs, the analysis can be performed quickly using the already-prepared data, thereby maintaining high reliability without incurring excessive processing time delays.
Solution Approach 2:
The patent creates copies of function execution information from multiple operating systems and stores them in a centralized storage medium. This copying approach allows the original OS operations to continue uninterrupted while the copied data is available for analysis, ensuring that the reliability of abnormality detection is maintained without causing time loss to the actual system operations being analyzed.
Data Source
AI summary
A non-transitory computer-readable recording medium stores therein an analysis program for causing a computer to execute a process including: acquiring information capable of identifying functions in operation which is obtained by a sampling by a plurality of operating systems at each first time interval with respect to programs in operation; totaling a number of pieces of the acquired information for each function; generating time-series data indicating the number of pieces of the information at each second time interval for the function whose number of pieces of the information satisfies a prescribed condition; analyzing a causal relationship between the functions based on the time-series data; and outputting an analysis result of the causal relationship between the functions.