&lt;title&gt;Expert network development environment for automating machine fault diagnosis&lt;/title&gt;

Adair, Kristin Lynn; Levis, Alan P.; Hruska, S.I.

doi:10.1117/12.235940

Cited by 2 publications

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Anomaly detection in distributed systems [10,13,11,15] detect when a failure has occurred, while the goal of Net-Poirot is to find the entity responsible for the failure. Inference and Trace-Based Algorithms [4,2,31,12] either require (a) data not locally available to the client at runtime, (b) knowledge/inference of the probability distribution of failure on each device in the system, (c) high resource consumption at runtime, or (d) knowledge/inference of application dependence on the different network/service devices. Each of these requirements raises the barrier of adoption, as compared to NetPoirot.…”

Section: Related Workmentioning

confidence: 99%

Taking the Blame Game out of Data Centers Operations with NetPoirot

Arzani

Ciraci

Loo

et al. 2016

Proceedings of the 2016 ACM SIGCOMM Conference

View full text Add to dashboard Cite

Today, root cause analysis of failures in data centers is mostly done through manual inspection. More often than not, customers blame the network as the culprit. However, other components of the system might have caused these failures. To troubleshoot, huge volumes of data are collected over the entire data center. Correlating such large volumes of diverse data collected from different vantage points is a daunting task even for the most skilled technicians. In this paper, we revisit the question: how much can you infer about a failure in the data center using TCP statistics collected at one of the endpoints? Using an agent that captures TCP statistics we devised a classification algorithm that identifies the root cause of failure using this information at a single endpoint. Using insights derived from this classification algorithm we identify dominant TCP metrics that indicate where/why problems occur in the network. We validate and test these methods using data that we collect over a period of six months in the Azure production cloud.

show abstract

Section: Related Workmentioning

confidence: 99%

Taking the Blame Game out of Data Centers Operations with NetPoirot

Arzani

Ciraci

Loo

et al. 2016

Proceedings of the 2016 ACM SIGCOMM Conference

View full text Add to dashboard Cite

show abstract

“…Using the knowledge table created by the team of experts, an expert network was developed [2]. The expert network for GC fault diagnosis consists of four layers of nodes: Figure 1 to an expert network is shown in Figure 3.…”

Section: Expert Network For Gc Fault Diagnosismentioning

confidence: 99%

“…The underlying form of the ystem described is a hybrid intelligence system called an exp i rt network. This hybrid method incorporates rule-based knowledge, training of weights in the rule-base using artificial neural network type techniques, and adaptive network structure refinement [2] [8] [14]. The system successfully recognizes faulty chromatograms from a variety of features detected in the graph and diagnoses the most likely cause of instrument failure.…”

Section: Introductionmentioning

confidence: 99%

Alternate path reasoning in intelligent instrument fault diagnosis for gas chromatography

Adair

Hruska²,

Elling³

Proceedings IEEE International Joint Symposia on Intelligence and Systems

View full text Add to dashboard Cite

elligent instrument fault diagnosis is addressed in this orks, a hybrid technique which sed expert systems with neural style training. One of the most difficult aspects ment fault diagnosis is developing an appropriate rk. Beginning with an initial rts, a more accurate representation can be found using example data. ning alternate paths of reaand incorporating them into the expert network is oses interaction and cotended to be used with stance of the expert to incorporate knowledge disfrom the data into the intelligent diagnosis tool. this methodology are conducted within the problem chromatography. Pereficacy of automating the asoning into the diagnostic 1 introduction

show abstract

<title>Expert network development environment for automating machine fault diagnosis</title>

Cited by 2 publications

References 0 publications

Taking the Blame Game out of Data Centers Operations with NetPoirot

Taking the Blame Game out of Data Centers Operations with NetPoirot

Alternate path reasoning in intelligent instrument fault diagnosis for gas chromatography

Contact Info

Product

Resources

About