A Performance Measurement and Analysis Method for Infotainment System 2020-01-1347
The car infotainment system provides various functions such as navigation, DMB, and external device interworking. In the future, complex new technologies such as remote diagnosis and control, V2X communication, integrated display, and autonomous driving interworking should be reflected. Also, even after mass production, such as smartphones, software must be constantly updated to apply bug fixes and new functions. However, as the internal software modules work together in a complicated way, system performance problems tend to occur due to changes in the software, and if they are not managed from the development stage, more costs must be paid for improvement over time. Moreover, after mass production, the scope of improvement will be limited. Therefore, software must be verified and optimized from time to time to ensure system performance quality from the minimum development process.
Generally, various tools are used to verify software performance. Currently, top, perf, strace, gdb, DLT, and bootchart are mainly used on Linux-based platforms. In addition, there is a lack of intuitive statistical information to analyze and judge the overall performance of the system. In addition, it is not easy to have an analysis environment because of the demanding conditions such as changing system settings, installing additional packages, and modifying software code for detailed performance analysis. In this paper, we introduce “Guider”, a tool that can measure and analyze system performance easily by providing intuitive and detailed performance statistics information at runtime without using software code.