Abstract
This paper describes the implementation of a ScanOnce algorithm in SQL for quick association rule mining and the development of a data mining infrastructure JetGrid. The architecture of JetGrid is designed to be compatible with lower-level grid mechanisms since it is to operate on top of Relational Grid Monitoring Architecture (R-GMA) provided by European Data Grid (EDG). JetGrid for quick knowledge discovery was preliminarily prototyped and it is extensible, using an object-oriented design that was coded in C++. Mining agents will be staged to one or more computing elements on the EDG. They will mine data delivered using GridFTP and/or Globus transport mechanism. Some initial experimental results are presented, which provide some indication of the performance of the JetGrid data mining infrastructure on top of EDG.
Original language | English |
---|---|
Pages | 811-814 |
DOIs | |
Publication status | Published - 7 Dec 2003 |
Event | International Conference on Grid and Cooperative Computing, GCC2003 - Shanghai, China Duration: 7 Dec 2003 → 10 Dec 2003 https://link.springer.com/book/10.1007/b97163 |
Conference
Conference | International Conference on Grid and Cooperative Computing, GCC2003 |
---|---|
Country/Territory | China |
City | Shanghai |
Period | 7/12/03 → 10/12/03 |
Internet address |