Experimental evaluation of the DECOS fault-tolerant communication layer

Abstract

This paper presents an experimental evaluation of the fault-tolerant communication (FTCOM) layer of the DECOS integrated architecture. The FTCOM layer implements different agreement functions that detect and mask errors sent either by one node using replicated communication channels or by redundant nodes. DECOS facilitates a move from a federated to an integrated architecture which means that non-safety and safety-related applications run on the same hardware infrastructure and use the same network. Due to the increased amount of data caused by the integration, the FTCOM is partly implemented in hardware to speed up packing and unpacking of messages. A cluster of DECOS nodes is interconnected via a time-triggered bus where transient faults with varying duration are injected on the bus. The goal of the experiments is to evaluate the fault-handling mechanisms and different agreement functions of the FTCOM layer. © Springer-Verlag Berlin Heidelberg 2007.