Timing Analysis

Timing-Analysis tools have recently entered industrial practice and are in routine use in the aeronautics and automotive industry. Precision of the results is good, efficiency is tolerable, and usability needs improvement.

All sectors concerned with Embedded Systems need Compilation Technology, WCET estimations are relevant for all industrial sections using hard real-time systems. Therefore, industrial sectors in this case include avionics, automotive, defence and some areas where control systems are applied. Especially in the automotive and the aeronautical domains, there is a need to have precise knowledge on the worst-case timing behaviour of safety critical software. This need is underlined by the fact that the worst-case timing of large parts of the software used within the new Airbus A380 has been analyzed using AbsInt’s aiT.

Both the aeronautics and the automotive industries follow a similar trend to integrated architectures, aeronautics to IMA (Integrated Modular Avionics), automotive to a component architecture developed by the AUTOSAR consortium. This transition at latest will require timing analysis as an integrated component in the development process.
However, only the availability of precise timing analyses does not fulfil industrial needs. Since the code of safety critical applications is typically generated by a compiler, the compiler should also be aware of worst-case timings. Currently, this is not the case leading to the unacceptable situation that, whenever it is detected that an application does not meet its real-time constraints, manual code transformation, recompilation and timing analysis need to be done repeatedly. The burden of timing analysis and optimization will be taken away from the human designer by the approach proposed at Dortmund.

Compilation

Mainly: Embedded software, semiconductor and system houses
Specifically: audio processing, video processing and data streaming applications in the TV, Set Top box, DVD player and recorder, mobile, base stations, printer and disk drive markets.

Efficiency of embedded software, in particular the efficiency of memories, is relevant for high-speed embedded systems. This includes multimedia and network applications. It is expected that most mobile devices will provide some kind of multimedia processing. Currently, most of the handheld devices feature processors with on chip memories. However, none of the industrial compilers support the automatic utilization of the on chip scratchpad memories. Moreover, no industrial tool exists for performing architectural level-exploration. Currently, all decisions regarding the optimal size of the on chip memories are taken by the system design through ad-hoc methods. Relevant industrial sectors include consumer electronics and telecommunications.

(c) Artist Consortium, All Rights Reserved - 2006, 2007, 2008, 2009

Réalisation Axome - Création de sites Internet