DocumentCode
3260348
Title
Parallel processors and an approach to the development of inference engine
Author
Prasad, Vinod B.
Author_Institution
Dept. of Electr. & Comput. Eng., Bradley Univ., Peoria, IL, USA
fYear
2005
fDate
23-25 May 2005
Firstpage
300
Lastpage
305
Abstract
The reliable and fault tolerant computers are key to the success to aerospace, and communication industries where failures of the system can cause a significant economic impact and loss of life. Designing a reliable digital system, and detecting and repairing the faults are challenging tasks in order for the digital system to operate without failures for a given period of time. The paper presents a new and systematic software engineering approach of performing fault diagnosis of digital systems, which have employed multiple processors. The fault diagnosis model is based on the classic PMC model to generate data obtained on the basis of test results performed by the processors. The PMC model poses a tremendous challenge to the user in doing fault analysis on the basis of test results performed by the processors. This paper will perform one fault model for developing software. The effort has been made to preserve the necessary and sufficient.
Keywords
inference mechanisms; parallel processing; software fault tolerance; aerospace industries; communication industries; digital systems; fault diagnosis model; inference engine; parallel processor; reliable fault tolerant computers; systematic software engineering; Aerospace industry; Communication industry; Digital systems; Engines; Fault detection; Fault diagnosis; Fault tolerant systems; Performance evaluation; Software engineering; Testing;
fLanguage
English
Publisher
ieee
Conference_Titel
Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing, 2005 and First ACIS International Workshop on Self-Assembling Wireless Networks. SNPD/SAWN 2005. Sixth International Conference on
Print_ISBN
0-7695-2294-7
Type
conf
DOI
10.1109/SNPD-SAWN.2005.58
Filename
1434904
Link To Document