| 摘要 |
A method for tracing a program with a relational database is provided to improve performance of the program by tracing function calls, searching program paths, and solving bottleneck with a parent function and call stack information of a specific time point. First trace data, which is used for tracing a program, of a function call is extracted from the program during execution of the program and is stored to a part of a field of a relation database. Second trace data is extracted by tracing the remaining part of the field from the stored first trace data of the function call after the program is terminated, and is stored to the remaining part of the field. The field includes an activation ID(10), a function ID(20), starting/keeping times(30,40) of the activation ID, initial child/sibling function activation IDs(50,60) of the activation ID, a next activation ID(70), and a parent function activation ID(80) of the activation ID. |
