Management of Real-Time Data Consistency and Transient Overloads in Embedded Systems

This thesis addresses the issues of data management in embedded systems’software. The complexity of developing andmaintaining software has increasedover the years due to increased availability of resources, e.g., more powerfulCPUs and larger memories, as more functionality can be accommodated usingthese resources.In this thesis, it is proposed that part of the increasing complexity canbe addressed by using a real-time database since data management is oneconstituent of software in embedded systems. This thesis investigates whichfunctionality a real-time database should have in order to be suitable forembedded software that control an external environment. We use an enginecontrol software as a case study of an embedded system.The findings are that a real-time database should have support for keepingdata items up-to-date, providing snapshots of values, i.e., the values are derivedfrom the same system state, and overload handling. Algorithms are developedfor each one of these functionalities and implemented in a real-time database forembedded systems. Performance evaluations are conducted using the databaseimplementation. The evaluations show that the real-time performance isimproved by utilizing the added functionality.Moreover, two algorithms for examining whether the system may becomeoverloaded are also outlined; one algorithm for off-line use and the secondalgorithm for on-line use. Evaluations show the algorithms are accurate andfast and can be used for embedded systems.