Error detection/correction and fault detection/recovery – Data processing system error or fault handling – Reliability and availability
Reexamination Certificate
1999-10-25
2002-11-19
Baderman, Scott (Department: 2184)
Error detection/correction and fault detection/recovery
Data processing system error or fault handling
Reliability and availability
C714S038110, C717S125000, C717S127000
Reexamination Certificate
active
06484276
ABSTRACT:
TECHNICAL FIELD
This invention relates to system dependability testing and, more particularly, to such testing via fault injection.
BACKGROUND OF THE INVENTION
Fault injection arrangements and techniques are known in the art. For the most part, prior known fault injection arrangements have limited flexibility regarding accommodation of hardware and software platforms, programs, and fault injection techniques. Indeed, fault injection is a necessity when testing the robustness of a system to unintended or unexpected events, because such events are often extremely difficult to produce through use of traditional testing techniques. See for example, an article by N. P. Kropp, P. J. Koopman and D. P. Siewiorek entitled “Automated robustness testing of off-the-shelf software components”,
Proceedings
28
th
International Symposium on Fault-Tolerant Computing
, (FTCS-28), pages 231-239, Munich Germany, June 1998, for one such prior fault injection tool. Notwithstanding, a significant obstacle to the task of systematic testing of system dependability is still the lack of flexible, easy-to-use fault injection tools.
SUMMARY OF THE INVENTION
The problems and limitations of prior fault injection tools are overcome in a unique fault injection tool based on an object-oriented architecture in which identical or similar functionality is abstracted out into so-called base classes, and is implemented in an object-oriented program language. This results in a fault injection tool having greater flexibility, ease and portability in realizing the basic functionality of the fault injection process including fault injection, workload generation and data collection.
Specifically, the basic functionality of the fault injection process is abstracted into three base classes, namely, a fault injector, a workload generator and a data collector. A control class performs configuration and management of the objects that are instantiated from the base classes. The control class also implements a graphical user interface. For each base class there is a corresponding core class that performs control and management of a so-called associated “plugin”. Each of the core classes can be implemented as a single class or as a pair of distributed classes. For example, the fault injector (FI) core class can be implemented as a single FI core class or as a FI/FIRemote pair. Similarly, the workload generator (WG) core class can be implemented as a single WG core class or as a WG/WGRemote pair. Likewise, the data collector (DC) core class can be implemented as a single DC core class or as a DC/DCRemote pair. If a core class is implemented as a pair, the FI, WG or DC object controls operation of the FIRemote, WGRemote or DCRemote object, respectively. For each core class, the associated plugin performs the actual functionality. A plugin is a dynamically loaded object that can be linked with the object instantiated from the core class without recompilation of the core class. Each plugin includes al least a corresponding base class and, possibly, hierarchical derived custom classes from the base class. The hierarchical derived custom classes facilitate software reuse and facilitate the creation of plugins. Many actions performed by the plugins are identical or similar for a wide range of fault injection, workload and data collection processes. These identical and similar actions are implemented in the abstract base classes. Intermediate classes can be derived from the base classes, and additional intermediate classes or the final end classes are derived from these intermediate classes.
REFERENCES:
patent: 4759019 (1988-07-01), Bently et al.
patent: 4999837 (1991-03-01), Reynolds et al.
patent: 5671352 (1997-09-01), Subrahmaniam et al.
patent: 6405363 (2002-06-01), Carlson et al.
Mei-Chen Hsueh et al., IEEE Computer, “Fault Injection Techniques and Tools”, Apr. 1997, vol. 30, Issue 4, pp. 75-82.*
Amana Cibele Apolinario Rosa et al., OOPSLA'98 Workshop #13, Vancouver, Canada, “Using a Reflective Architecture to Validate Object-oriented Applications by Fault Injection”, Oct. 18, 1998.*
N. P. Kropp et al., “Automated Robustness Testing of Off-the-Shelf Software Components”, Proceedings28th International Symposium on Fault-Tolerant Computing, (FTCS), pp. 231-239, Munich, Germany Jun. 1998.
Singh Navjot
Tsai Timothy
Baderman Scott
Lohn Joshua
Lucent Technologies - Inc.
Stafford Thomas
LandOfFree
Method and apparatus for providing extensible... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for providing extensible..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for providing extensible... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2988854