Browse Prior Art Database

An autonomic system and method for automated rule composition based on artifact contents

IP.com Disclosure Number: IPCOM000127362D
Original Publication Date: 2005-Aug-25
Included in the Prior Art Database: 2005-Aug-25
Document File: 4 page(s) / 28K

Publishing Venue

IBM

Abstract

Suggested an autonomic system and method for automated rule composition based on artifact contents

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 26% of the total text.

Page 1 of 4

An autonomic system and method for automated rule composition based on artifact contents

The core of this invention disclosure is the self-optimizing of the rule sets that can be used by multiplicity of applications to self-optimize their own operations, for example the Generic Log Adapter. These unique self-optimizing and self-configuring features provided here allow a user-friendly and intuitive approach to the rule set writing. For example parsing rules for conversion from native log formats to the Common Base Event format while at the same time doing run time optimizations to reduce the performance overhead of using regular expressions. Optimizations are possible at both design time (rule creation) and runtime (collective rule execution). Data conversion into a common format is a key requirement for autonomic computing. This invention assumes significance because it enabled such conversion through user friendly rule construction and induces optimal runtime rules execution. While regular expressions have been used for extraction of data, such a behavior of a collection of regular expressions operating on the same data block over the complete input stream has not be implemented before.

To address the complexities of applying regular expression techniques (i.e., rule sets) for the construction of generic and optimal rule sets for easy migration of native log formats to the, for example but not limited to, Common Base Event format, we have provided autonomic mechanism that could be used, but not limited to, as part of the design of the Generic Log Adapter. Some of the features are useful in design time when the rules are being created while others are useful in runtime when the rule set is optimized for low performance overhead. These mechanisms greatly decrease the runtime performance overhead while at the same time keeping the design time experience simple and user-friendly. In other words, the user is able to write simple rules, which the runtime then manipulates, based on recorded characteristics and patterns observed at runtime, to provide a customized optimization of the conversion process for that particular native format or possible combination of native formats.

This disclosure provides novel techniques for composition of parsing rules. Example of preferred embodiment of this disclosure is described in terms of the Autonomic Computing Generic Log Adapter (GLA) and Common Base Event (CBE). Details of these techniques are outlined below:

1. Regular expression guessing

As part of the Rule Builder tool provided with the Generic Log Adapter, there is a feature to visually highlight a portion of a data block in the native log format and derive a regular expression that will match that block. The regular expression can be further modified by highlighting the actual portion within the previously highlighted data block portion that represents the information to be extracted to populate the corresponding CBE property for which this...