Call-Trace Classification Rules for Distributed System Root Cause Analysis

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

The increasing complexity of distributed computing systems has led to management and administration challenges, including significant inefficiencies and computational overheads, making traditional approaches to automated management and administration impractical, especially in diagnosing operational problems and failures within these systems.

Innovation Solution

The use of call-trace-classification rules generated from automatically labeled call traces to facilitate root-cause analysis of distributed-application operational problems and failures, which helps identify specific services and service failures, providing valuable information for managers and administrators.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Extent of automation

If traditional automated management and administration approaches are used in distributed computing systems, then management functionalities can be automated, but the complexity and computational overhead increase significantly making the systems impractical to manage

Engineering Contradiction:

Improveautomation of management and administrationVSAvoidcomplexity of management system

Core Design Contradiction:

Extent of automationVSDevice complexity

Solution Approach 1:

The patent segments the complex management system into separate functional modules: call-trace collection, call-trace classification, root-cause analysis, and diagnosis tools. Each module handles specific tasks independently, reducing overall system complexity while maintaining automation capabilities. The call-trace classification module specifically segments analysis into multiple diagnostic layers to manage complexity.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces call-trace classification rules as an intermediary layer between raw call traces and root-cause analysis. These classification rules simplify the processing of complex distributed system data by pre-defining diagnostic categories and thresholds, enabling automated management without requiring complex real-time analysis algorithms.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Measurement precision

If comprehensive root-cause analysis is performed on all call traces, then diagnostic accuracy improves, but computational overhead and analysis time increase significantly

Engineering Contradiction:

Improvediagnostic accuracyVSAvoidanalysis time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent performs preliminary classification of call traces using pre-generated call-trace-classification rules before conducting comprehensive root-cause analysis. This preliminary action filters and organizes call traces into relevant categories based on diagnostic criteria, enabling accurate analysis of only the most pertinent traces and reducing overall analysis time.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent applies different analysis depths and diagnostic methods to different call-trace categories based on their local characteristics. Critical call traces requiring detailed analysis are processed with higher precision, while less critical traces are analyzed more superficially, optimizing the balance between diagnostic accuracy and computational resources.

Inventive Principle:

Principle #3Local quality

3Adaptability or versatility

If complex diagnostic algorithms are implemented, then root-cause analysis capability improves, but the system becomes more difficult to implement and maintain

Engineering Contradiction:

Improveroot-cause analysis capabilityVSAvoidease of implementation

Core Design Contradiction:

Adaptability or versatilityVSEase of manufacture

Solution Approach 1:

The system generates and maintains its own diagnostic capabilities through automated call-trace-classification-rule generation. The diagnosis tools self-improve by learning from analyzed call traces and updating classification rules, reducing the need for manual implementation and maintenance of complex diagnostic algorithms while enhancing root-cause analysis capability.

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS11880271B2Automated methods and systems that facilitate root cause analysis of distributed-application operational problems and failures

Publication Date: 2024.01.23 VMWARE INC

AI summary

The current document is directed to methods and systems that employ call traces collected by one or more call-trace services to generate call-trace-classification rules to facilitate root-cause analysis of distributed-application operational problems and failures. In a described implementation, a set of automatically labeled call traces is partitioned by the generated call-trace-classification rules. Call-trace-classification-rule generation is constrained to produce relatively simple rules with greater-than-threshold confidences and coverages. The call-trace-classification rules may point to particular services and service failures, which provides useful information to distributed-application and distributed-computer-system managers and administrators attempting to diagnose operational problems and failures that arise during execution of distributed applications within distributed computer systems. Call-trace-classification rules that are useful in multiple diagnoses are maintained as diagnosis tools for future diagnoses.