Hosted by Utah State University Libraries
Advancement of Computing on Large Datasets via Parallel Computing and Cyberinfrastructure
Abstract
dc:description.abstractLarge datasets require high processing power to compute, high-speed network connections to transmit, or high storage capacity to archive. With the advent of the internet, many in the science community and the public at large are faced with a need to manage, store, transmit and process large datasets in an efficient fashion to create value for all concerned. By example, large environmental researchers analyze large map data to extract hydrologic information from topography. However, processing these data and other tasks is hard – sometimes impossible – in minimal resource environments such as desktop systems. This dissertation demonstrates novel approaches and programming algorithms that address several issues associated with large datasets. We present a novel virtual memory system that can process large map data which are too big to fit in memory in raster-based calculations. Then, we introduce parallel computer algorithms that reduce large data to smaller, more easily managed forms; and finding the patterns inside of this smaller representation of large data efficiently in large computer systems. We need also software services on the internet to access large remote computer systems. We present a software service named HydroGate that abstracts away many details and complexities involved in the use of large remote computer systems, including authentication, authorization, and data and job management.
Degree
thesis:*- Name thesis:degree_name
- Doctor of Philosophy (PhD)
- Level thesis:degree_level
- Dissertation
- Discipline thesis:degree_discipline
- Computer Science
- Grantor dc:publisher
- Hosted by Utah State University Libraries
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Yildirim, Ahmet Artu
- Contributors dc:contributor
-
- Daniel Watson
- David Tarboton
- Vladimir Kulyukin