Date of Award

2015

Document Type

Thesis

Degree Name

Bachelors

Department

Natural Sciences

First Advisor

Shipman, Steven

Keywords

Autofit, Amazon EC2, Computer Science

Area of Concentration

Chemistry

Abstract

Recent developments in instrumentation have made it possible to collect broadband rotational spectra far faster than those spectra can be assigned. As such, we have been working to develop automated assignment algorithms so that the analysis can catch up with the data acquisition. The Autofit program has made strides in this direction, but it is still quite slow on spectra with high line densities, such as those collected near room temperature. Given that the Autofit algorithm is highly parallelizable, we have used Amazon’s EC2 webservice to run a modified version of Autofit simultaneously across a large number of cores, allowing us to obtain results in a fraction of the time normally required by a typical desktop computer. This thesis describes how Autofit was modified to run on EC2 and presents some benchmark results.

Share

COinS