A proposal to apply inductive logic programming to self-healing problem in grid computing: How will it work?
✍ Scribed by Mariza Ferro; Antonio Roberto Mury; Bruno Schulze
- Publisher
- John Wiley and Sons
- Year
- 2011
- Tongue
- English
- Weight
- 317 KB
- Volume
- 23
- Category
- Article
- ISSN
- 1532-0626
- DOI
- 10.1002/cpe.1714
No coin nor oath required. For personal study only.
✦ Synopsis
SUMMARY
As computation systems get extremely large and complex, failure diagnosis becomes even more complex. To cope with this ever increasing complexity of managing heterogeneous systems—such as grids and nowadays clouds—systems should manage their own behavior themselves. This vision of self‐managing systems also referred to as autonomic computing (AC) aims to allow systems to recover themselves from various failures or malfunctions. This is known as self‐healing (SH) and is one of the requirements of AC. However, dealing with these complex failure scenarios is always an open challenge. Dealing with this challenge requires prediction and control through a number of automated learning and proactive actions. In this work, we present the usage of a relational learning method known as inductive logic programming, for prediction and root casual analysis, and the development of an SH component. Copyright © 2011 John Wiley & Sons, Ltd.