Inductive learning for fault diagnosis

Berenji, H.R.; Ametha, Jayesh; Vengerov, David

doi:10.1109/fuzz.2003.1209453

Cited by 19 publications

(12 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Similar result was observed by Schroeder and Gibson [29]. several studies [1,37,39] have examined system logs to identify causal events that lead to failures. Correlation between the workload intensity and the failure rate in real systems has been pointed out in many studies [5,23,25,19,3].…”

Section: Related Worksupporting

confidence: 70%

Quantifying event correlations for proactive failure management in networked computing systems

2010

Journal of Parallel and Distributed Computing

View full text Add to dashboard Cite

Section: Related Worksupporting

confidence: 70%

Quantifying event correlations for proactive failure management in networked computing systems

2010

Journal of Parallel and Distributed Computing

View full text Add to dashboard Cite

“…Similar result was observed by Schroeder and Gibson [29]. several studies [8,37,38] have examined system logs to identify causal events that lead to failures. Correlation between the workload intensity and the failure rate in real systems was pointed out in many studies [14,24,26,20,12].…”

Section: Related Worksupporting

confidence: 70%

“…Figure 7(a) presents the results in predicting hardware and software caused failures with different prediction algorithms. According to the figure, NN (8) and AR(8) performed the best among them. This is because of the learning capacity of these two approaches.…”

Section: Sensitivity To Prediction Algorithmsmentioning

confidence: 99%

Exploring event correlation for failure prediction in coalitions of clusters

2007

Proceedings of the 2007 ACM/IEEE Conference on Supercomputing

149

118

View full text Add to dashboard Cite

In large-scale networked computing systems, component failures become norms instead of exceptions. Failure prediction is a crucial technique for self-managing resource burdens. Failure events in coalition systems exhibit strong correlations in time and space domain. In this paper, we develop a spherical covariance model with an adjustable timescale parameter to quantify the temporal correlation and a stochastic model to describe spatial correlation. We further utilize the information of application allocation to discover more correlations among failure instances. We cluster failure events based on their correlations and predict their future occurrences. We implemented a failure prediction framework, called PREdictor of Failure Events Correlated Temporal-Spatially (hP), which explores correlations among failures and forecasts the time-between-failure of future instances. We evaluate the performance of hP in both offline prediction of failure by using the Los Alamos HPC traces and online prediction in an institute-wide clusters coalition environment. Experimental results show the system achieves more than 76% accuracy in offline prediction and more than 70% accuracy in online prediction during the time from

show abstract

“…Oil filter differential pressure kPa (psig) u 6 Post filter oil pressure kPa (psig) u 7 Pre filter oil pressure kPa (psig) u 8 Oil rifle pressure kPa (psig) u 9 -u 10 LB/RB boost pressures kPa (psig) u 11 -u 14 All banks IMT 1C (1F) u 15 Coolant pressure kPa (psig) u 16 Coolant temperature C (F) u 17 Rail pressure kPa (psig) u 18 Battery Y-output, u-inputs, LB-left bank, RB-right bank, F-front, R-rear, IMT-intake manifold temperature, EGT-exhaust gas temperature. Comparing the performance of the RGS to the best accuracy of 89.7% with 8 sensors reported in [27], it can be seen that performing simultaneous sensor selection and model development is indeed beneficial to improving the performance of the system.…”

Section: Article In Pressmentioning

confidence: 99%

“…Multi-sensor data fusion seeks to increase accuracy by exploiting complementary information, while at the same time increase reliability by exploiting the redundancy provided by different sensors. Many attempts towards achieving this goal can be found in literature [3][4][5][6][7][8][9].…”

Section: Introductionmentioning

confidence: 99%