XCEPTION - A TECHNIQUE FOR THE EXPERIMENTAL EVALUATION OF DEPENDABILITY IN MODERN COMPUTERS

Citation
J. Carreira et al., XCEPTION - A TECHNIQUE FOR THE EXPERIMENTAL EVALUATION OF DEPENDABILITY IN MODERN COMPUTERS, IEEE transactions on software engineering, 24(2), 1998, pp. 125-136
Citations number
49
Categorie Soggetti
Computer Science Software Graphycs Programming","Engineering, Eletrical & Electronic","Computer Science Software Graphycs Programming
ISSN journal
00985589
Volume
24
Issue
2
Year of publication
1998
Pages
125 - 136
Database
ISI
SICI code
0098-5589(1998)24:2<125:X-ATFT>2.0.ZU;2-Z
Abstract
An important step in the development of dependable systems is the vali dation of their fault tolerance properties. Fault injection has been w idely used for this purpose, however with the rapid increase in proces sor complexity, traditional techniques are also increasingly more diff icult to apply. This paper presents a new software implemented fault i njection and monitoring environment, called Xception, which is targete d for the modern and complex processors. Xception uses the advanced de bugging and performance monitoring features existing in most of the mo dern processors to inject quite realistic faults by software, and to m onitor the activation of the faults and their impact on the target sys tem behavior in detail. Faults are injected with minimum interference with the target application. The target application is not modified, n o software traps are inserted, and it is not necessary to execute the target application in special trace mode (the application is executed at full speed). Xception provides a comprehensive set of fault trigger s, including spatial and temporal fault triggers, and triggers related to the manipulation of data in memory. Faults injected by Xception ca n affect any process running on the target system (including the kerne l), and it is possible to inject faults in applications for which the source code is not available. Experimental results are presented to de monstrate the accuracy and potential of Xception in the evaluation of the dependability properties of the complex computer systems available nowadays.