Abstract: Methods that use marking, leveling and linking (“MLL”) processes to identify problems and dynamically correlate events recorded in various log files generated for a use-case of an application are described. The marking process determines fact objects associated with the use-case from events recorded in the various log files, database dumps, captured user actions, network traffic, and third-party component logs in order to identify non-predefined problems with running the application in a distributed computing environment. The MLL methods do not assume a predefined input format and may be used with any data structure and plain log files. The MLL methods present results in a use-case trace in a graphical user interface. The use-case trace enables human users to monitor and troubleshoot execution of the application. The use-case trace identifies the types of non-predefined problems that have occurred and points in time when the problems occurred.

Abstract: Methods that use marking, leveling and linking (“MLL”) processes to identify problems and dynamically correlate events recorded in various log files generated for a use case of an application are described. The marking process determines fact objects associated with the use-case from events recorded in the various log files, database dumps, captured user actions, network traffic, and third-party component logs in order to identify non-predefined problems with running the application in a distributed computing environment. The MLL methods do not assume a predefined input format and may be used with any data structure and plain log files. The MLL methods present results in a use-case trace in a graphical user interface. The use-case trace enables human users to monitor and troubleshoot execution of the application. The use-case trace identifies the types of non-predefined problems that have occurred and points in time when the problems occurred.