Automated Culprit Thread Identification in CPU Lockups

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional methods for determining the source of software crashes are error-prone, time-consuming, and often inaccurate, as they rely on manual comparison of stack traces, which can lead to incorrect attribution of crash causes and delayed customer support.

Innovation Solution

An automated end-to-end system that analyzes crash reports by identifying culprit threads and generating signature back traces, using a combination of module identification and graphical user interfaces to provide accurate and timely insights into software crash causes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual comparison of stack traces is used to determine crash source, then human judgment and flexibility are applied, but the process becomes time-consuming and error-prone

Engineering Contradiction:
Improveaccuracy of crash source identificationVSAvoidtime required for crash analysis
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent replaces the manual mechanical process of comparing stack traces with an automated computer-based system that uses algorithms to analyze crash reports, extract stack traces, and identify culprit threads automatically, eliminating human intervention while maintaining or improving accuracy

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system enables self-service crash analysis by automatically processing crash reports and generating identification of the culprit thread without requiring human analysts, allowing the system to serve itself in analyzing its own crash data

Inventive Principle:
Principle #25Self-service

2Reliability

If manual analysis methods are used, then detailed human inspection can be performed, but incorrect attribution of crash causes occurs

Engineering Contradiction:
Improveaccuracy of crash cause attributionVSAvoidspeed of customer support response
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent replaces manual human inspection with automated computational analysis that systematically processes crash reports and reliably identifies culprit threads, eliminating human errors in attribution while accelerating the support response process

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system provides automated feedback by analyzing crash reports and immediately identifying the culprit thread, creating a closed-loop process that continuously improves reliability through systematic analysis rather than subjective human judgment

Inventive Principle:
Principle #23Feedback

3Measurement precision

If comprehensive stack trace analysis is performed manually, then thorough examination is possible, but the process becomes tedious and complex

Engineering Contradiction:
Improveaccuracy of crash source identificationVSAvoidcomplexity of analysis process
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent replaces the complex manual process of analyzing comprehensive stack traces with an automated system that handles the complexity internally through algorithms, presenting a simplified output to users while maintaining thorough analysis capabilities

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system extracts only the essential information needed for crash source identification from the comprehensive stack trace data, separating the complex analysis process from the simple output presentation, thereby reducing perceived complexity while maintaining analysis thoroughness

Inventive Principle:
Principle #2Taking out (Extraction)

Data Source

PatentUS11099971B2Determination of a culprit thread after a physical central processing unit lockup
Publication Date: 2021.08.24 VMWARE INC
  • US11099971B2 patent drawing
  • US11099971B2 patent drawing
  • US11099971B2 patent drawing

AI summary

An automated end-to-end analysis of customer service requests is disclosed. A core dump is received, wherein the core dump corresponds to a customer service request regarding a crash of a computer system. A processor automatically analyzes the core dump to determine if a pcpu lockup of the computer system is due to a software issue. Provided the pcpu lockup of the computer system is due to the software issue, the processor determines which thread is a culprit thread responsible for the pcpu lockup of the computer system.