Fault Data Analytics Using Decision Tree for Fault Detection

Tran, Ha Manh; Nguyen, Son C.; Le, Son T.; Vu, Quy Tran

doi:10.1007/978-3-319-26135-5_5

Cited by 2 publications

(5 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We have evaluated the performance of the proposed approach and CART decision tree on both log event data and context-aware data collected from the cloud system. The experimental results show that this approach has achieved the high accuracy score of 85% compared to the previous approach [2] with the accuracy score of 71% on average.…”

Section: Discussionmentioning

confidence: 88%

“…An observation reveals that providing more log records collected from the real system can make the decision tree larger and more accurate. The previous study [2] has also built the decision tree based on log events and obtained the maximum accuracy score of 0.71. Our method considers both log events and context-aware log records for improving accuracy.…”

Section: Discussionmentioning

confidence: 99%

“…A fault manager needs sufficient storage, bandwidth and processing power to perform several complicated operations, such as search and share fault resources, receive update on fault datasets, monitor and evaluate suspected events. Our previous study [2] has proposed the fault detection system in communication networks and distributed systems. While the previous study uses software bug reports to evaluate suspected events, this study aims at exploiting both context-aware data and event to evaluate suspected events.…”

Section: System Architecturementioning

confidence: 99%

“…We have applied the ASF-BDT method for event correlation [17] and the CART decision tree for event inspection [2]. The correlation method contains multiple filtering processes: starting with log event data, applying with simple and complex filtering methods, finally returning event correlated data.…”

Section: Event Correlation and Inspectionmentioning

confidence: 99%

“…Among several studies of event analysis, the study of Tran et al [2] has applied the classification and regression decision trees (CART) [3] for evaluating the severity level of events automatically, thus enabling system administrators to decide whether further steps are needed for detecting faults. The approach focuses on constructing a CART decision tree based on the features of existing events and then using the tree to determine the severity level of occurring events.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Exploiting Context-Aware Event Data for Fault Analysis

Nguyen

et al. 2016

REV-JEC

Self Cite

View full text Add to dashboard Cite

Fault analysis in communication networks and distributed systems is a difficult process that heavily depends on system administrator’s experience and supporting tools. This process usually requires analytic techniques and several types of event data including log events, debug messages, trace obtained from these systems to investigate the root cause of faults. This paper introduces an approach of exploiting context-aware data and classification technique for improving this process. This approach uses both event data and context-aware data including CPU load, memory, processes, temperature, status to train a decision tree, and then applies the tree to assess suspected events. We have implemented and experimented the approach on the OpenStack cloud computing system with the Hadoop computing service and MELA event collection system. The experimental results reveal that the accuracy score of the approach reaches 85% on average. The paper also includes detailed analysis for the results.

show abstract

Section: Discussionmentioning

confidence: 88%

Section: Discussionmentioning

confidence: 99%

Section: System Architecturementioning

confidence: 99%

Section: Event Correlation and Inspectionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations