Date of Award

11-21-2014

Document Type

Thesis

Degree Name

Computer Science, MS

First Advisor

Hai Jiang

Committee Members

Hung-Chi Su; Jeff Jenness; XiuZhen Huang

Call Number

LD 251 .A566t 2014 Q46

Abstract

The MapReduce programming model has been criticized for a long time for its lack of flexibility to apply on many difficult scientific computations. Currently, several approaches try to conduct a more flexible MapReduce framework, but some of them are required to run on a particular platform, others have a lack of support to GPU environments. In this thesis, we present MR-Tree, a customizable GPU MapReduce framework. In MR-Tree, users can configure each task's data I/O behavior, so computations that have strong data dependency will no longer hold back the MapReduce runtime. Moreover, MR-Tree also features configurable workflow. Users can arrange the task nodes in several different ways to handle complicated MapReduce jobs. We will discuss the customization for MR-Tree to handle intricate, iterative, long-running and regular MapReduce applications.

Rights Management

Creative Commons Attribution 4.0 International License
This work is licensed under a Creative Commons Attribution 4.0 International License.

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.