Redundant parallel data transfer schemes for the grid environment

In this paper, we proposed dynamic co-allocation schemes for parallel data transfer in a grid environment, which copes up with highly inconsistent network performances of the servers. We have developed an algorithm using circular queue, with which, the data transfer tasks are allocated onto the servers in duplication. It is further improved to maintain the ordered delivery of data blocks, to suit such applications, like streaming. The schemes nicely tackle even when the link to servers under consideration is broken or become idle. We used Globus toolkit for our framework and utilized the partial copy feature of GridFTP. We compared our schemes with the existing schemes and the preliminary results show notable improvement in overall completion time of data transfer. Keywords : datagrid, parallel data transfer, coallocation

• Type: Conference
• Year: 2006
• Where: ACSW
• Authors: Raghuvel S. Bhuvaneswaran, Yoshiaki Katayama, Naohisa Takahashi
• Keywords: ACSW 2006, ACSW 2007, Data Transfer Tasks, Dynamic Co-allocation Schemes, Parallel Data