Performance Monitoring Features in EPOS
ResumoIncorporating performance monitoring capabilities to embedded environments, especially on critical systems such as Cyber-Physical Systems, requires a negligible intrusion to ensure the CPS environment performance and data quality. The monitoring design must be tailored to fit the system’s needs instead of being limited to a single monitoring approach. In this paper, we extend a monitoring framework to encompass three monitoring approaches: Periodic-, Execution-flow-, and Job-based, focusing on the evaluation of overhead, latency, and jitter for each of them. We have implemented and evaluated these approaches over the Monitoring Framework design, where none of them presented an average impact on the system execution time higher than 0.3%. While the Job-based monitoring presented better results in terms of impact over the tasks execution and memory consumption, the Execution-flow-based monitoring presented better results for jitter, both on the impact over task execution time and the monitoring latency.
Palavras-chave: Measurement, Time-frequency analysis, Data integrity, Instruments, Memory management, Focusing, Interference, Performance Monitoring, Monitoring Framework, Time-triggered, Execution-flow-based, Job-based
HORSTMANN, Leonardo Passig; HOFFMANN, José Luís Conradi; FRÖHLICH, Antônio Augusto. Performance Monitoring Features in EPOS. In: SIMPÓSIO BRASILEIRO DE ENGENHARIA DE SISTEMAS COMPUTACIONAIS (SBESC), 11. , 2021, Evento Online. Anais [...]. Porto Alegre: Sociedade Brasileira de Computação, 2021 . p. 16-23. ISSN 2237-5430.