Unified Trace Data Structure for Distributed System Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In complex distributed systems, tracing operations across multiple computer systems of different types is challenging due to incompatibility and differing operation methods, making it difficult to obtain a complete picture of application performance and identify issues.
Innovation Solution
An end-to-end tracing method that enables simultaneous tracing on multiple systems, using a common trace format to combine data from various systems, with correlation objects to merge and analyze traces, and an administration system to manage and implement these traces across Java and proprietary systems like ABAP.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If performance traces are obtained on multiple different computer systems with different computer environments, then a complete picture of application operation can be obtained, but the systems are incompatible and handle operations differently, making it difficult to generate a complete analysis
Solution Approach 1:
The patent creates a universal trace data structure that can accommodate multiple computer environments (mainframe, midrange, PC) with different operating systems and hardware architectures. The data structure uses standardized field definitions that can represent trace information from any system type, allowing a single analysis tool to handle traces from diverse sources without requiring system-specific processing logic.
Solution Approach 2:
The patent transforms trace data from different computer environments into a common parameter format. By defining standardized fields (such as trace record identifier, timestamp, function name, parameter names and values) that can represent the essential information from any system, the patent changes the parameters of trace data to be compatible across platforms, enabling unified analysis despite underlying system differences.
2Productivity
If a computer application runs on multiple different systems of different types, then power and efficiency of operation are increased, but the distributed application operation complicates the analysis process if a problem occurs
Solution Approach 1:
The patent segments the distributed application's trace data into standardized, identifiable units that can be independently processed and then reassembled. Each system generates trace records with unique identifiers that segment the overall application execution into system-specific portions, making it easier to locate and analyze problems in specific segments without being overwhelmed by the complexity of the entire distributed system.
Solution Approach 2:
The patent introduces an intermediary standardized trace data structure that mediates between different computer environments. This intermediate representation layer translates traces from various systems into a common format, allowing analysis tools to work with a unified data structure rather than dealing directly with the complexity of multiple incompatible system formats.
3Loss of information
If standard trace formats are used in each system, then trace data can be collected from multiple systems, but the trace data from different systems cannot be easily combined due to format incompatibility
Solution Approach 1:
The patent merges trace data from multiple systems into a unified data structure. By defining a common format that combines essential elements from different system-specific trace formats (such as combining mainframe trace identifiers with PC-style parameter structures), the patent enables seamless integration of trace data from diverse sources into a single analyzable dataset.
Data Source
AI summary
A method and apparatus for tracing operations in multiple computer systems. An embodiment of a method includes running a computer application, where the computer application operates on multiple computer systems, including a first system and a second system. The first system and the second system have different computer environments. A trace of the computer application is performed, with performance of the trace including enabling a first trace of the first system and enabling a second trace of the second system. Trace data is collected from the first system and the second system during the operation of the application, and the trace data from the first system and the trace data from the second system are combined.