Software based Fault Tolerance may allow the use of COTS digital electronics in building a highly reliable computing system for spacecraft. In this work we present the results of a fault injection campaign we conducted on the Duplex Framework (DF). The DF is a software developed by the UCLA group [1], [2] that allows to run two copies (or replicas) of the same program on two different nodes of a commercial off-the-shelf (COTS) computer cluster. By the means of a third process (comparator) running on a different node that constantly monitors the results computed by the two replicas, the DF is able to restart the two replica processes if an inconsistency in their computation is detected. In order to test the reliability of the DF we wrote a simple fault injector that injects faults in the virtual memory of one of the replica process to simulate the effects of radiation in space. These faults occasionally cause the process to crash or produce erroneous outputs. For this study we used two different applications, one that computes an encryption of a input file using the RSA algorithm, and another that optimizes the trade-off between time spent and the fuel consumption for a low-thrust orbit transfer. But the DF is generic enough that any application written in C or Fortran could be used with little or no modification of the original source code. Our results show the potential of such approach in detecting and recovering from radiation induced random errors.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Fault Injection Campaign for a Fault Tolerant Duplex Framework


    Contributors:


    Publication date :

    2007-03-01


    Size :

    10862699 byte





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Fault injection for dependability validation of fault-tolerant computing systems

    Arlat, J. / Crouzet, Y. / Laprie, J.C. | Tema Archive | 1989


    Fault-Tolerant Components versus Fault-Tolerant Systems

    Muller, A. / Plankensteiner, M. / VDI/VDE-Gesellschaft Mess und Automatisierungstechnik | British Library Conference Proceedings | 2002



    Fault diagnostics and fault tolerant control

    Zhou, D.H. / Frank, P.M. | IEEE | 1998


    Fault-Tolerant Control

    Ding, Ruqi / Cheng, Min | Springer Verlag | 2024