Grid Application Fault Diagnosis Using Wrapper Services and Machine Learning

Abstract

With increasing size and complexity of Grids manual diagnosis of individual application faults becomes impractical and timeconsuming. Quick and accurate identification of the root cause of failures is an important prerequisite for building reliable systems. We describe a pragmatic model-based technique for application-specific fault diagnosis based on indicators, symptoms and rules. Customized wrapper services then apply this knowledge to reason about root causes of failures. In addition to user-provided diagnosis models we show that given a set of past classified fault events it is possible to extract new models through learning that are able to diagnose new faults. We investigated and compared algorithms of supervised classification learning and cluster analysis. Our approach was implemented as part of the Otho Toolkit that 'serviceenables' legacy applications based on synthesis of wrapper service.

Toward Energy‐Aware Scheduling Using Machine Learning

“…Detecting failures in resources and applications, and also the root of failures, has become a very interesting area where inductive learning methods are also applied. Approaches such the one presented by Hofer and Fahringer [12] shows an application-specific fault diagnosis based on indicators, symptoms, and rules. For this approach, two techniques have been used: a supervised classification to find the reason of the failure and clustering techniques to find what failures are the result of the same cause.…”

Finding associations in Grid monitoring data

“…A classification of grid faults and the application of supervised classification and cluster techniques to enhance grid fault detection is provided by [7].…”