Uma Arquitetura Altamente Disponível Aplicada a Sistemas de Controle Embutidos de Tempo Real
Abstract
An architecture based on controller redundancy for increased embedded real-time systems safety and availability is presented. Two COTS controllers process the same inputs and the computed data is compared as a means to detect errors. When an error is detected, a diagnosis routine tries to identify its locality. This may take some control cycles, which violates the real-time property of the controlled application; however, this situation is allowed to happen for the embedded systems we consider as long as it does not cause the failure of the controlled process.
References
Cunha, J.C. et al. A study of failure models in feedback control systems. In: International Conference on Dependable Systems and Networks, Goteburg, 2001. Proceedings. Los Alamitos: IEEE Computer Society, Jul. 2001.
Jones, C.T.; Bryan, L.A. Programmable controllers – concepts and applications. USA: International Programmable Controls, 1983. 329p.
Laprie, J.C. Dependability of computer systems: from concepts to limits. In: IFIP International Workshop on Dependable Computing and its Applications. Johannesburg, South Africa, 1998. Proceedings. [S.l.: s.n. , Jan. 1998.
Pradhan, D.K.; Banerjee, P. Fault-tolerant multiprocessor and distributed systems: principles. In: PRADHAN, D. K., Fault-tolerant computer system design. Upper Saddle River: Prentice Hall, 1996. 550p. cap.3, p.135-235.
QNX Developer’s Network. Disponível em [link] Acesso em: abril de 2003.
Suri, N.; Walter, C.J e Hugue, M.M. Advances in Ultra-Dependable distributed systems. cap.1. Los Alamitos: IEEE Computer Society, 1995. 467p.
