Comparison of physical and software-implemented fault injection techniques

Arlat, Jean; Crouzet, Yves; Karlsson, Johan; Folkesson, Peter; Fuchs, Emmerich; Leber, Günther

doi:10.1109/tc.2003.1228509

Cited by 147 publications

(55 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Perturbation, repeatability, monitoring-time resolution, controllability are recognized as some of the most significant properties for fault injection tools [10], [11], [12], [13]; while designing both fault injection tools and experiments, typically relevant efforts are spent to properly address them. As examples in the field of SWIFI, we briefly mention the tools ORCHESTRA [14], XCEPTION [15] and LOKI [16] and an experiment [17] performed using the FTAPE tool.…”

Section: Guidelines and Good Practices For Swifimentioning

confidence: 99%

Trustworthy Evaluation of a Safe Driver Machine Interface through Software-Implemented Fault Injection

Ceccarelli

Bondavalli

Iovino

2009

2009 15th IEEE Pacific Rim International Symposium on Dependable Computing

View full text Add to dashboard Cite

Abstract-Experimental evaluation is aimed at providing useful insights and results that constitute a confident representation of the system under evaluation. Although guidelines and good practices exist and are often applied, the uncertainty of results and the quality of the measuring system is rarely discussed. To complement such guidelines and good practices in experimental evaluation, metrology principles can contribute in improving experimental evaluation activities by assessing the measuring systems and the results achieved. In this paper we present the experimental evaluation by software-implemented fault injection of a safe train-borne Driver Machine Interface (DMI), to evaluate its behavior in presence of faults. The measuring system built for the purpose and the results obtained on the assessment of the DMI are scrutinized along basic principles of metrology and good practices of fault injection. Trustfulness in results has been estimated satisfactory and the experimental campaign has shown that the safety mechanisms of the DMI correctly identify the faults injected and that a proper reaction is executed.

show abstract

Section: Guidelines and Good Practices For Swifimentioning

confidence: 99%

Trustworthy Evaluation of a Safe Driver Machine Interface through Software-Implemented Fault Injection

Ceccarelli

Bondavalli

Iovino

2009

2009 15th IEEE Pacific Rim International Symposium on Dependable Computing

View full text Add to dashboard Cite

show abstract

“…As these are also slow, the experimental approach of fault injection has been used instead to test real systems under realistic faults. Fault injection can be HW [12,3,4] or SW implemented [2-6, 13, 14]. In HW-implemented injection, faults are injected physically by electromagnetic interference and radiation [12] or through the circuit pins [3,4].…”

Section: Related Workmentioning

confidence: 99%

“…Fault injection can be HW [12,3,4] or SW implemented [2-6, 13, 14]. In HW-implemented injection, faults are injected physically by electromagnetic interference and radiation [12] or through the circuit pins [3,4]. Although these inject real hardware faults that can reach all locations, they lack flexibility and are difficult to operate and control.…”

Section: Related Workmentioning

confidence: 99%

“…On the contrary, SW-implemented fault injection (SWIFI) overcomes most of these drawbacks by injecting realistic faults using software methods. SWIFI achieves higher properties of repeatability, controllability (in space and time), reproducibility, non-intrusiveness and efficacy [12]. Generally, in SWIFI, transient faults are injected by adding traps or replacing instructions, either at preruntime or at runtime.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Understanding the Effects of Data Corruption on Application Behavior Based on Data Characteristics

Stefanakis

Nagarajan

Cintra

2015

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. In this paper, the results of an experimental study on the error sensitivities of application data are presented. We develop a portable software-implemented fault-injection (SWIFI) tool that, on top of performing single-bit flip fault injections and capturing their effects on application behavior, is also data-level aware and tracks the corrupted application data to report their high-level characteristics (usage type, size, user, memory space location). After extensive testing of NPB-serial (7.8M fault injections), we are able to characterize the sensitivities of data based on their high-level characteristics. Moreover, we conclude that application data are error sensitive in parts; depending on their type, they have distinct and wide less-sensitive bit ranges either at the MSBs or LSBs. Among other uses, such gained insight could drive the development of sensitivity-aware protection mechanisms of application data.

show abstract

“…These methodologies, which require a prototype to introduce the faults into the system, are further divided into Hardware Implemented Fault Injection (HWIFI) and Software Implemented Fault Injection (SWIFI) [13] [14].…”

Section: Introductionmentioning

confidence: 99%

Speeding-up model-based fault injection of deep-submicron CMOS fault models through dynamic and partially reconfigurable FPGAS

Martinez¹,

de²

View full text Add to dashboard Cite

ResumenActualmente, las tecnologías CMOS submicrónicas son básicas para el desarrollo de los modernos sistemas basados en computadores, cuyo uso simplifica enormemente nuestra vida diaria en una gran variedad de entornos, como el gobierno, comercio y banca electrónicos, y el transporte terrestre y aeroespacial. La continua reducción del tamaño de los transistores ha permitido reducir su consumo y aumentar su frecuencia de funcionamiento, obteniendo por ello un mayor rendimiento global. Sin embargo, estas mismas características que mejoran el rendimiento del sistema, afectan negativamente a su confiabilidad. El uso de transistores de tamaño reducido, bajo consumo y alta velocidad, está incrementando la diversidad de fallos que pueden afectar al sistema y su probabilidad de aparición. Por lo tanto, existe un gran interés en desarrollar nuevas y eficientes técnicas para evaluar la confiabilidad, en presencia de fallos, de sistemas fabricados mediante tecnologías submicrónicas.Este problema puede abordarse por medio de la introducción deliberada de fallos en el sistema, técnica conocida como inyección de fallos. En este contexto, la inyección basada en modelos resulta muy interesante, ya que permite evaluar la confiabilidad del sistema en las primeras etapas de su ciclo de desarrollo, reduciendo por tanto el coste asociado a la corrección de errores. Sin embargo, el tiempo de simulación de modelos grandes y complejos imposibilita su aplicación en un gran número de ocasiones.Esta tesis se centra en el uso de dispositivos lógicos programables de tipo FPGA (Field-Programmable Gate Arrays) para acelerar los experimentos de inyección de fallos basados en simulación por medio de su implementación en hardware reconfigurable. Para ello, se extiende la investigación existente en inyección de fallos basada en FPGA en dos direcciones distintas: i) se realiza un estudio de las tecnologías submicrónicas existentes para obtener un conjunto representativo de modelos de fallos transitorios y permanentes que deben ser tenidos en cuenta, y se analiza hasta que punto estos modelos pueden emularse por medio de FPGA, y ii) para cada modelo de fallo considerado, se describen diversos procedimientos alternativos para realizar su emulación, evaluando la aceleración obtenida en cada caso.FADES (F PGA-based Framework for the Assessment of the D ependability of E mbedded S ystems) es el prototipo desarrollado para ilustrar la viabilidad de esta metodología. Los experimentos realizados por medio de FADES se han centrado en comprobar la validez de los resultados obtenidos y mostrar la aceleración alcanzable por comparación con una herramienta de inyección de fallos reconocida. Finalmente, se deducen las principales ventajas e inconvenientes de la aproximación propuesta, así como su utilidad para la evaluación de la confiabilidad de sistemas empotrados.i Resum En l'actualitat, les tecnologies CMOS submicròniques són bàsiques pel desenvolupament dels moderns sistemes basats en computadors, els quals simplifiquen enormement la nostra vida diària e...

show abstract

Comparison of physical and software-implemented fault injection techniques

Cited by 147 publications

References 40 publications

Trustworthy Evaluation of a Safe Driver Machine Interface through Software-Implemented Fault Injection

Trustworthy Evaluation of a Safe Driver Machine Interface through Software-Implemented Fault Injection

Understanding the Effects of Data Corruption on Application Behavior Based on Data Characteristics

Speeding-up model-based fault injection of deep-submicron CMOS fault models through dynamic and partially reconfigurable FPGAS

Contact Info

Product

Resources

About