Cooperative Storage Subsystem for Automatic Data Transfer in Grid Computing Environment

초록

The computing sites have their own specific softwares installed and researchers or engineers may need several softwares to work in cooperation. A researcher may login one computing site to use a necessary software and work, then the person logout the computing site and login another site to use next necessary software. In this scenario, the person’s input data and output data should be transferred to each computing site. If necessary softwares are dispersed in each computing site, the person’s data should be transferred as many as the number of computing sites. If recomputing of previous software is necessitated which is detected by a result of the next software, this would implicate the person much more data transfers. Our goal is automatic data transfer among storage subsystems to service cooperatively in the Grid computing environ-ment, and this data service would relieve the researchers from tedious manual data transferring. In our recent knowledge, there has not been any approach similar to our goal in Grid computing, since proposed storage subsystem supply remote storage space to an application as virtually ‘local’ disks.