Award Date
5-2010
Degree Type
Thesis
Degree Name
Master of Science in Computer Science
Department
Computer Science
First Committee Member
Jan B. Pedersen, Chair
Second Committee Member
Kazeem Taghva
Third Committee Member
Yoohwan Kim
Graduate Faculty Representative
Henry Selvaraj
Number of Pages
74
Abstract
Information Visualization has been used as an effective method in transforming textual information into a visual form that enables users to effectively communicate and understand the information. MPI (Message Passing Interface) usually involves a large amount of data, which necessitates exploring innovative ideas to visualize such data effectively.
In this thesis, we implement a graph visualization tool called MPROV to effectively visualize communication patterns of message passing using log files from MPI applications. The tool provides several interaction techniques to effectively reduce the amount of data displayed on the screen. We have also developed protocol conformance checking, which verifies the correctness of the communication by using a protocol specification file containing a list of constraints. Finally, we test our application with different log files containing a list of communication patterns and also at different strictness level that the messages should conform to based on the protocol constraints.
Keywords
Computer networks; Data transmission systems; Digital communications; Human-computer interaction; Information visualization
Disciplines
Computer Sciences | Graphics and Human Computer Interfaces
File Format
Degree Grantor
University of Nevada, Las Vegas
Language
English
Repository Citation
Sadasivan, Arunkumar, "A Visualization approach for message passing in parallel computing systems" (2010). UNLV Theses, Dissertations, Professional Papers, and Capstones. 328.
http://dx.doi.org/10.34917/1564110
Rights
IN COPYRIGHT. For more information about this rights statement, please visit http://rightsstatements.org/vocab/InC/1.0/