DBSitter: um sistema inteligente para gerencimaneto de SGBD
AUTOR(ES)
Adriana Paula Carneiro Cavalcanti
DATA DE PUBLICAÇÃO
2005
RESUMO
By analyzing the daily work of Data Base Administrators (ABD), we perceived that a large amount of problems (e. g. performance tuning, security and physical dimensioning) occur repetitively. Nonetheless, the professionals normally forget the best solution for them, especially when working under pressure. Remembering the correct solution may take a long time and the service quality can be compromised. Most existent products for managing and monitoring Database Management Systems (DBMS) have the main features expected for tools that manage DBMS, namely: continuous monitoring of DBMS resources, automatic fault detection and correction, and problems prediction. However, some limitations have been found in those products: some solutions are specific for an even DBMS or can not be extended with the inclusion of new problems or solutions. Another limitation is their inability to use acquired knowledge in solving previous similar problems to solve the new ones. They just solve or predict problems that are identical to a previously catalogued problem that is already stored in their repository. In order to solve these limitations we have developed the DBSitter: a multiagent system for managing and monitoring Database Management Systems (DBMS). DBSitter is a flexible, extensible and multiplatform system that assists the ABD in their decision making by suggesting adequate solutions for the problems and by taking advantage of acquired knowledge when previous similar problems were solved. Moreover, DBSitter innovates in using two Artificial Intelligence techniques: Case Based Reasoning and Intelligent Agents. The first experiments done with the system yielded promising results and validated the solution architecture, while showing that the main expected functionalities were present
ASSUNTO(S)
intelligent agents raciocÃnio baseado em casos agentes inteligentes managing and monitoring dbms gerenciamento e monitoraÃÃo de sgbd ciencia da computacao inteligÃncia artificial case based reasoning artificial intelligence
Documentos Relacionados
- DBSitter-AS: um Framework Orientado a Agentes para ConstruÃÃo de Componentes de Gerenciamento AutÃnomo para SGBD
- Sistema inteligente para elaborar um projeto de perfuração de um poço de petroleo
- Um sistema difuso inteligente para avaliar informações de usuários na Internet
- Um sistema inteligente baseado em ontologia para apoio ao esclarecimento de dúvida
- Desenvolvimento de um sistema de manutenção inteligente embarcado