S. Poledna Abstract: This paper discusses the problem of message-based interprocess communication in embedded real-time control systems and its appropriate operating system support. Based on state message semantics a novel optimization approach for interprocess communication is introduced. This approach provides high efficiency with respect to execution time and memory requirements. Its high efficiency is achieved by exploiting a priori available knowledge about the static system structure in an embedded system for optimizations. The operating system can thus provide a general purpose state message communication mechanism. But instead of using a general purpose implementation the actual implementation is tailored for a given application scenario. A case study based on a real application in the context of automotive electronics is carried out. This study shows the superior performance of the presented optimization strategy compared to event-messages and unoptimized state messages. Furthermore, it shows that systematic application of message-based interprocess communication becomes possible with the optimized communication mechanism. The optimized state message mechanism presented here has influenced the design of the operating system ERCOS [17] which will be used by Bosch for a next generation of automotive control units. for a given application scenario. A case study based on a real application in the context of automotive electronics is carried out. This study shows the superior performance of the presented optimization strategy compared to event-messages and unoptimized state messages. Furthermore, it shows that systematic application of message-based interprocess communication becomes possible with the optimized communication mechanism. The optimized state message mechanism presented here has influenced the design of the operating system ERCOS [17] which will be used by Bosch for a next generation of automotive control units.
[1]
Michael Wolfe,et al.
Communication Analysis for Multicomputer Compilers
,
1994,
IFIP PACT.
[2]
Arthur J. Bernstein,et al.
Concurrency in programming and database systems
,
1993
.
[3]
Sheng Cheng,et al.
Scheduling algorithms for hard real-time systems: a brief survey
,
1989
.
[4]
Norihisa Suzuki,et al.
Shared memory multiprocessing
,
1992
.
[5]
Hermann Kopetz,et al.
Event-Triggered Versus Time-Triggered Real-Time Systems
,
1991,
Operating Systems of the 90s and Beyond.
[6]
Alan Burns,et al.
Hard Real-Time Scheduling: The Deadline-Monotonic Approach
,
1991
.
[7]
Gianluigi Castelli,et al.
EOS: a real-time operating system adapts to application architectures
,
1995,
IEEE Micro.
[8]
J. Schiemann,et al.
ERCOS: An Operating System for Automotive Applications
,
1996
.
[9]
M. Raynal,et al.
How to find his way in the jungle of consistency criteria for distributed shared memories (or how to escape from Minos' labyrinth)
,
1993,
1993 4th Workshop on Future Trends of Distributed Computing Systems.
[10]
Harvey M. Deitel,et al.
An introduction to operating systems
,
1984
.
[11]
Fred B. Schneider,et al.
Implementing fault-tolerant services using the state machine approach: a tutorial
,
1990,
CSUR.