Dependability analysis of a high-speed network using software-implemented fault injection and simulated fault injection

Abstract

This paper presents a dependability study of high-speed, switched Local Area Networks (LANs) using Myrinet as an example testbed (with theoretical speeds of 2.56 Gbps). The study uses results of two fault injection methods, simulated fault injection and softwareimplemented fault injection (SWIFI), to analyze the application-level impact of transient faults injected into the network interface hardware. These results include a number of errors such as dropped or corrupt messages, host interface or host resets, and local or remote host interface hangs. The paper presents the study in two parts: First, the results from the SWIFI method in the real system are used as a basis to validate the simulation and identify the major factors leading to di erences between the methods. A comparison between the two injection methods shows that they agree for 83% of the fault injections. The results, however, vary greatly depending on the fault type considered. The study presents an analysis of the e ects of varying workload intensity, host platform, and interface function targeted by the injection. An example of this analysis is to show that the function targetted has a signi cant impact on the fault activation rate. Finally, the study identi es two mechanisms by which faults may propagate from the interface to other parts of the network; in one example, this propagation caused the interface's host computer to reboot while another caused a remote interface in the network to hang.

Dependency Analysis in Distributed Systems using Fault Injection: Application to Problem Determination in an e-commerce Environment

“…Software fault injection tools greatly outnumber the hardware tools. Some examples are NFTAPE from the University of Illinois [11], Ferrari developed at the University of Texas at Austin [9] and Xception from the University of Coimbra in Portugal [3]. A good overview of the field of fault injection is to be found in [5].…”

Designing fault-injection experiments for the reliability of embedded systems

“…There are a number of programs and tools for efficient fault injection [2,7,10,22,24,26,28,31,35,43,45,49,55,56]. There are efforts to model fault propagation and effect [5,7,12,17,19,22,23,27,28,30,32,36,38,39,40,41,44,45,47,50,52]. Coverage is a popular topic, and the meaning of coverage can vary from simple detection to complete system recovery and reconfiguration [1,2,3,4,5,10,11,12,15,16,17,18,20,21,27,29,…”

Contribución a la validación de sistemas complejos tolerantes a fallos en la fase de diseño. Nuevos modelos de fallos y técnicas de inyección de fallos.

“…Otra aplicación interesante se presenta en [Stott et al 1997, Stott et al 1998]. En este caso se aplica SWIFI para el análisis de la Confiabilidad de Myrinet, una red comercial de alta velocidad.…”

“…La inyección se lleva a cabo mediante una aplicación que modifica las instrucciones que se ejecutan en la parte de comunicaciones, escribiendo en la memoria de la interfaz. En [Stott et al 1997, Stott et al 1998] se muestran los resultados de dicha aplicación. En [Stott et al 1998], además, se comparan con los obtenidos al aplicar inyección de fallos mediante simulación sobre un modelo del mismo sistema.…”

“…En [Stott et al 1997, Stott et al 1998] se muestran los resultados de dicha aplicación. En [Stott et al 1998], además, se comparan con los obtenidos al aplicar inyección de fallos mediante simulación sobre un modelo del mismo sistema. En este caso, la herramienta de inyección utilizada es DEPEND (véase el apartado 3.2.5).…”