Probabilistic diagnosis of performance faults in large-scale parallel applications.
Ignacio LagunaDong H. AhnBronis R. de SupinskiSaurabh BagchiTodd GamblinPublished in: PACT (2012)
Keyphrases
- posterior probability
- fault diagnosis
- model based diagnosis
- bayesian networks
- multiple faults
- fault detection
- generative model
- probabilistic model
- fault detection and diagnosis
- fault model
- root cause
- real world
- diagnostic reasoning
- parallel processing
- small scale
- parallel implementation
- real life
- expert systems
- neural network
- fault models
- web scale
- parallel computation
- graphical models
- parallel execution
- artificial intelligence
- fault isolation