In partial fulfillment of the Requirements for the Degree of
Master of Science
Leiming Shao
will defend his thesis
Heuristic and Visualization of Scheduling
Communication in an Overloaded Network
Abstract
Data Staging is an
important data management issue for distributed computer systems. Most Previous
works are concentrated on the static version of data staging problem. In this
thesis, we introduce a dynamic scheduling algorithm to address the issues of
distributing and storing both repository data and continuously generated data
over numerous geographically dispersed locations though an overloaded network,
where not all requests can be satisfied.
The our scheduling algorithm would use limited storage
capacities to satisfy as many request as possible before their deadlines. In
our algorithm, the low quality data copies are used to try to satisfy with the
requests that could not be satisfied by the high quality data due to limited
network resource. Intermediate data in
temporary storage space will be cached as data source to help requests before
these data are purged from the storage.
In addition, a visualization program is built based
on the extendible Graph library developed using Java language.
Date: Wednesday, December
4, 2002
Time: 11:00 AM
Place: 550-PGH
Faculty, students, and the general public are invited.
Thesis Advisor: Dr. Stephen Huang