Methodology for Performance Analysis of Distributed Knowledge-Based Systems
Nataliia A. Kulykovska, Артур Валентинович Тіменко, Svitlana Hrushko, Stepan Skrupsky · 2021
In this paper, the performance of distributed knowledge-based systems is the time spent by an application to execute messaging functions.To analyze the performance of distributed knowledge-based systems, direct performance measurement with tracing of web service function calls is used.The main advantage of this approach is the preservation of complete information about the interaction of processes in the trace, which makes it possible to analyze the performance of web services in detail.The identification of performance problems based on the analysis of the collected data is determined by the selected criterion.The paper proposes an expert methodology for solving the performance problem.The scientific novelty of the work lies in the presentation of the model of the problem of the performance of web services, as a set of actions of processes, certain combinations of which, under certain conditions, can lead to an identifiable performance problem.