Support for end-to-end response-time and delay analysis in the industrial tool suite: Issues, experiences and a case study

Abstract

In this paper we discuss the implementation of the state-of-theart end-to-end response-time and delay analysis as two individual plug-ins for the existing industrial tool suite Rubus-ICE. The tool suite is used for the development of software for vehicular embedded systems by several international companies. We describe and solve the problems encountered and highlight the experiences gained during the process of implementation, integration and evaluation of the analysis plug-ins. Finally, we provide a proof of concept by modeling the automotive-application case study with the existing industrial model (the Rubus Component Model), and analyzing it with the implemented analysis plug-ins.

Designing end-to-end resource reservations in predictable distributed embedded systems

“…3 for details). These activation patterns are commonly used in the industrial control systems (Feiertag et al 2008;Mubeen et al 2013). …”

“…The rest of the tasks in the chain are activated by their predecessors. Such a chain is called a trigger chain (Mubeen et al 2013). The input of the chain corresponds to the data read by τ 1 from register Reg-1.…”

“…In this chain, each and every task is activated independently, i.e, no task is activated by its predecessor. Such a chain is called a data chain (Mubeen et al 2013). The periods of activation for tasks τ 1 , τ 2 and τ 3 are 8, 8 and 4 ms respectively.…”

“…Whereas, the reaction delay is important in the applications where the first reaction to the input is of value, e.g., in the button-to-reaction applications in the body electronics domain. We refer the reader to Feiertag et al (2008) and Mubeen et al (2013) for the calculations of end-to-end response times and delays in single-node embedded systems without resource reservations.…”

“…The age and reaction delays are also identified. Once again, we refer the reader to Feiertag Mubeen et al (2013) for the calculations of the age and reaction delays in distributed embedded systems without resource reservations.…”