Skip to Main Content
 

Global Search Box

 
 
 
 

Files

File List


osu1250594496.pdf (413.47 KB)

ETD Abstract Container

Abstract Header

Measuring and Improving the Potential Parallelism of Sequential Java Programs

Van Valkenburgh, Kevin
http://rave.ohiolink.edu/etdc/view?acc_num=osu1250594496

Abstract Details

2009, Master of Science, Ohio State University, Computer Science and Engineering.
There is a growing need for parallel algorithms and their implementations, due to the continued rise in the use of multi-core machines. When trying to parallelize a sequential program, software engineers are faced with the problem of understanding and removing inhibitions to parallelism in the context of high-level program constructs that are used to implement modern software. This thesis describes a dynamic analysis to analyze potential parallelism in Java software, based on coarse-grain parallelism among method calls. We find that this automated analysis, when applied to 26 sequential Java programs, provides characterization and insights about potential parallelism bottlenecks. We also perform case studies on several of the analyzed programs, locating specific elements in the implementation that impede potential parallelism, and altering these elements to find greater potential parallelism. Our studies indicate that the dynamic analysis could provide valuable feedback to a software engineer in support of program understanding and transformations for parallelism.
Atanas Rountev, PhD (Advisor)
P. Sadayappan, PhD (Committee Member)
118 p.
Computer Science
Java; parallelism; multi-core; potential parallelism

Recommended Citations

Citations

  • Van Valkenburgh, K. (2009). Measuring and Improving the Potential Parallelism of Sequential Java Programs [Master's thesis, Ohio State University]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=osu1250594496

    APA Style (7th edition)

  • Van Valkenburgh, Kevin. Measuring and Improving the Potential Parallelism of Sequential Java Programs. 2009. Ohio State University, Master's thesis. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=osu1250594496.

    MLA Style (8th edition)

  • Van Valkenburgh, Kevin. "Measuring and Improving the Potential Parallelism of Sequential Java Programs." Master's thesis, Ohio State University, 2009. http://rave.ohiolink.edu/etdc/view?acc_num=osu1250594496

    Chicago Manual of Style (17th edition)

osu1250594496
588
© 2009, all rights reserved.
This open access ETD is published by The Ohio State University and OhioLINK.