Graphical Call Tree Filtering for Performance Root Cause Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Modern web applications face difficulties in tracing the root cause of poor-performing transactions due to complex reports and alerts, which do not adequately indicate the cause of issues in redundant systems with millions of transactions processed daily.
Innovation Solution
A method involving accessing a call tree for transactions, generating a graphical representation of the call tree based on performance properties, allowing users to graphically select routines, creating a filtered call tree, and generating a drill-down visualization to efficiently identify performance issues.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If complex reports and alerts are used to track transactions, then more information is provided about transaction status, but it becomes difficult to trace the root cause of poor-performing transactions
Solution Approach 1:
The patent segments the complex call tree into a graphical representation where routines are organized hierarchically with visual cues. The call tree is divided into nodes representing different routines, with graphical indicators showing performance properties. This segmentation allows users to see individual routine performance without being overwhelmed by the entire system complexity.
Solution Approach 2:
The patent adds a graphical dimension to the call tree data by representing routines visually with spatial relationships. Instead of flat text reports, the call tree is displayed as a hierarchical graph where vertical position indicates call depth and horizontal positioning shows routine relationships. Performance properties are represented through graphical attributes like color coding and visual markers, adding another dimension for analyzing transaction performance.
2Measurement precision
If detailed call tree information is displayed, then root cause analysis capability is improved, but user ability to quickly identify specific problematic routines is reduced
Solution Approach 1:
The patent uses color changes and graphical indicators to highlight problematic routines within the call tree. Routines with poor performance properties are marked with distinct visual cues such as color coding or special symbols. This allows users to quickly scan the graphical representation and identify problematic areas without manually analyzing each routine's detailed information, significantly reducing troubleshooting time while maintaining accurate root cause identification.
3Quantity of substance
If the entire call tree is analyzed, then comprehensive performance coverage is achieved, but the ability to focus on specific problematic areas is lost
Solution Approach 1:
The patent allows users to extract and focus on specific portions of the call tree by enabling selection of individual routines or groups of routines within the graphical representation. When users identify problematic areas through visual inspection, they can extract those specific routines for detailed analysis while setting aside the rest of the call tree. This extraction capability maintains comprehensive coverage when needed while enabling focused analysis of specific problematic areas, improving overall analysis efficiency.
Data Source
AI summary
In an embodiment, a method includes accessing a call tree for a transaction, wherein the call tree traces routines called during execution of the transaction. The method further includes generating a graphical representation of the call tree in relation to two or more performance properties. In addition, the method includes causing the graphical representation of the call tree to be displayed. Further, the method includes allowing a user to graphically select a group of routines from the graphical representation of the call tree. In addition, the method includes creating a filtered call tree comprising the graphically selected group of routines. Furthermore, the method includes generating a drill-down visualization of the filtered call tree. The method also includes causing the drill-down visualization to be displayed.