Dissertation > Excellent graduate degree dissertation topics show

Internet service fault diagnosis algorithm

Author: LiZuo
Tutor: ZuoXiRong
School: Beijing University of Posts and Telecommunications
Course: Computer Science and Technology
Keywords: Fault Management Service Troubleshooting Hidden Markov propagation model Viterbi decoding Active probes
CLC: TP393.06
Type: Master's thesis
Year: 2009
Downloads: 75
Quote: 0
Read: Download Dissertation

Abstract


With the development of Internet technology, Internet is gradually to a service-oriented architecture (Service Oriented Architecture, SOA) evolution. User-centric SOA, the goal is to provide users with a good service experience, allowing users at any time, any place to get good service. Fault management is very important for the service QoS assurance, and service will be unavailable or performance degradation caused by SLA (Service Level Agreement, Service Level Agreement) breach of contract, affecting the SP's reputation and cause economic losses. Therefore, SP needs service-oriented fault management mechanism, the service fails to quickly locate the fault and take appropriate measures to reduce service downtime and performance degradation time. This paper focuses on the Internet service fault management system and its related diagnostic algorithm is studied, research focus is on technical services based on probabilistic fault diagnosis, service fault propagation modeling, fault location algorithm analysis and improvement, HMM fault propagation, and Viterbi-based fault hypothesis reasoning. The main working paper contains the following aspects: (1) analysis of service failures caused by complex reasons, comparison of the current fault diagnosis technology, select Graph theory as theoretical tools of this study, the use of probabilistic model proposed HMM fault propagation model. HMM model by establishing a dependency between failures metastasis, simplifying the troubleshooting process. (2) was proposed based on hidden Markov process (Hidden Markov Model) fault propagation model (FPM) of the complete formal expression of a new generation of fault propagation model in the modeling process, to consider the transfer of the relationship between failure, while Consider transfer between faults and symptoms relationship. Transfers between failures is not visible to the outside world, that can not be directly identified the faulty component. Fault propagation through the HMM model for modeling the system uncertainty. (3) to obtain data on the network for real difficulty, the probe selected by the fault information entropy priori probability for the HMM with fault propagation model, the proposed one after the forward backward algorithm (Forward-backward) algorithm, the probe needle selection into the model parameters obtained in the process to make the model parameters obtained with the probe chose to become an organic whole. (4) Failure to diagnose the problem can be described as the observed symptoms are known under the premise of reasoning faulty component of the process. Through the information entropy gain maximum probe set minimum coverage standards, reasonable reduction in diagnostic probe generated network load. Reasoning part of the fault hypothesis presented incomplete information under the Viterbi N-best way to sort through, retaining partial function prior to the first N records, recursively calculated the global optimum, reasoning fault source.

Related Dissertations

  1. The Low Power Research of Adaptive Receiver of OFDM Digital Base-band,TN851
  2. Band Entropy Method and Its Application to Fault Diagnosis of Rolling Bearings,TH165.3
  3. Engine cold test and processing of data and application of multivariate correlation,U464
  4. Marine diesel engine condition monitoring and diagnostic system development,U664.121
  5. Slurry pipeline transportation system, process monitoring and fault diagnosis,F426.1
  6. Dedicated circuit test methods Research and Implementation,TN707
  7. Networked Control System Fault Diagnosis and Fault Tolerant Control,TP273
  8. Methods based on RDT FDF diagnostic systems for oil drilling accident,TE28
  9. Based on Blind separation of large wind turbine mechanical fault signal feature extraction,TK83
  10. Recloser based distribution automation system optimization design and research,TM76
  11. Based on Bayesian Networks motor fault diagnosis method,TM307.1
  12. Based on CLIPS CBM gathering process fault diagnosis expert system design,TP182
  13. Helicopter main gearbox fault diagnosis and failure prediction technology research,V267
  14. Smart appliances Fault Diagnosis System Research and Application,TP273
  15. Helicopter transmission chain fault diagnosis and health management system design and key technology research,V267
  16. Formation of collision between satellites and anti-collision technology,V448.2
  17. Electric transmission lines based fault diagnosis,TM755
  18. Integrated Intelligent Computing Based Fault Diagnosis of Analog Circuits,TN710
  19. Analog Circuit Fault Diagnosis Method and Its Application,TN710
  20. Based on Information Fusion Fault Diagnosis of Analog Circuits,TN710
  21. Fault Diagnosis Based on Data Mining Methods,TP311.13

CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer applications > Computer network > General issues > Computer networks, test , run
© 2012 www.DissertationTopic.Net  Mobile