Runtime support for reconfigurable real-time embedded systems

As the embedding environment becomes more and more complex so does the embedded system itself. An aspect of the complexity, the demand for robust and fault tolerant embedded solutions is ever increasing. Consequently the embedded system design and development face new challenges including modeling, representation, execution and implementation issues. These issues cannot be addressed adequately without a matching underlying system architecture. Model integrated computing (MIC) architectures relying on multi-aspect modeling and variety of models of computation are proven to be promising alternatives for designing and generating manageable complex computing systems for various application domains. With respect to implementation one of the substantial questions is how the MIC based (self-) adapting design can be mapped into a (reconfigurable) distributed hardware platform. The paper presents a novel middleware supporting execution of adapting/reconfigurable real-time programs. The main features of MIC and its relation to adaptivity and computation model based programming are overviewed. A minimal set of functionalities, which enables to construct real-time program structures serving as an integration framework for various models of computation is presented. The design principles of the runtime environment are summarized.

[1]  Ian T. Foster Automatic Generation of Self-Scheduling Programs , 1991, IEEE Trans. Parallel Distributed Syst..

[2]  Edward A. Lee,et al.  A framework for comparing models of computation , 1998, IEEE Trans. Comput. Aided Des. Integr. Circuits Syst..

[3]  Rajiv Gupta,et al.  A compact task graph representation for real-time scheduling , 1996, Real-Time Systems.

[4]  Gabor Karsai,et al.  The multigraph and structural adaptivity , 1993, IEEE Trans. Signal Process..

[5]  Gabor Karsai,et al.  A Configurable Visual Programming Environment: A Tool for Domain-Specific Programming , 1995, Computer.

[6]  David Lorge Parnas,et al.  On synchronization in hard-real-time systems , 1988, CACM.

[7]  Z. Papp,et al.  A multi-agent based modeling and execution framework for complex simulation, control and measuring tasks , 2000, Proceedings of the 17th IEEE Instrumentation and Measurement Technology Conference [Cat. No. 00CH37066].

[8]  Gabor Karsai,et al.  Model-Integrated Computing , 1997, Computer.

[9]  Stephen A. Edwards,et al.  Design of embedded systems: formal models, validation, and synthesis , 1997, Proc. IEEE.