Detail publikace

On Distribution and Impact of Fault Effects at Real-Time Kernel and Application Levels

Originální název

On Distribution and Impact of Fault Effects at Real-Time Kernel and Application Levels

Anglický název

On Distribution and Impact of Fault Effects at Real-Time Kernel and Application Levels

Jazyk

en

Originální abstrakt

Real-time operating systems are often utilized to simplify the design of embedded time/safety-critical applications. However, real-time embedded systems are sensitive to transient and other faults, each of which can lead to various errors at different system levels and can potentially result in the system failure. In the previous papers, a real-time kernel sensitivity to soft-errors was studied and the errors being induced were classified. In this paper, sensitivity to transient, permanent and firmware faults is analyzed on basis of data gathered from series of experiments over the uC/OS-II kernel. During each of the experiments, various dependability techniques (N-version programming, control-flow checking, watchdog tasks and processing of pairs) - and their combinations - were implemented into the kernel and various faults were injected into the kernel in order to study impact of their effects to both response timeliness and behavior correctness of the system.

Anglický abstrakt

Real-time operating systems are often utilized to simplify the design of embedded time/safety-critical applications. However, real-time embedded systems are sensitive to transient and other faults, each of which can lead to various errors at different system levels and can potentially result in the system failure. In the previous papers, a real-time kernel sensitivity to soft-errors was studied and the errors being induced were classified. In this paper, sensitivity to transient, permanent and firmware faults is analyzed on basis of data gathered from series of experiments over the uC/OS-II kernel. During each of the experiments, various dependability techniques (N-version programming, control-flow checking, watchdog tasks and processing of pairs) - and their combinations - were implemented into the kernel and various faults were injected into the kernel in order to study impact of their effects to both response timeliness and behavior correctness of the system.

BibTex


@inproceedings{BUT96940,
  author="Josef {Strnadel} and František {Slimařík}",
  title="On Distribution and Impact of Fault Effects at Real-Time Kernel and Application Levels",
  annote="Real-time operating systems are often utilized to simplify the design of embedded
time/safety-critical applications. However, real-time embedded systems are
sensitive to transient and other faults, each of which can lead to various errors
at different system levels and can potentially result in the system failure. In
the previous papers, a real-time kernel sensitivity to soft-errors was studied
and the errors being induced were classified. In this paper, sensitivity to
transient, permanent and firmware faults is analyzed on basis of data gathered
from series of experiments over the uC/OS-II kernel. During each of the
experiments, various dependability techniques (N-version programming,
control-flow checking, watchdog tasks and processing of pairs) - and their
combinations - were implemented into the kernel and various faults were injected
into the kernel in order to study impact of their effects to both response
timeliness and behavior correctness of the system.",
  address="IEEE Computer Society",
  booktitle="Proceedings of the 15th Euromicro Conference on Digital System Design: Architectures, Methods and Tools",
  chapter="96940",
  doi="10.1109/DSD.2012.90",
  edition="NEUVEDEN",
  howpublished="print",
  institution="IEEE Computer Society",
  year="2012",
  month="september",
  pages="272--279",
  publisher="IEEE Computer Society",
  type="conference paper"
}