Graduate Thesis Or Dissertation
 

Performance monitoring of parallel applications at large grain level

Öffentlich Deposited

Herunterladbarer Inhalt

PDF Herunterladen
https://ir.library.oregonstate.edu/concern/graduate_thesis_or_dissertations/t435gh777

Descriptions

Attribute NameValues
Creator
Abstract
  • This thesis is an attempt to create a methodology to analyze the performance of parallel applications on a wide variety of platforms and programming environments. First we determined the monitoring functions required to collect traces for accurate representation of the parallel application. We used the Extended Large Grain Data Flow (E L G D F) representation of an application to determine granularity and which monitoring functions should be inserted for sufficient feedback to application designer. The monitoring routines ( real time clock access procedures ) with a common interface were developed for the Sequent [superscript] TM multiprocessor machine and the C-Linda programming environment . We also developed an Execution Profile Analyzer( E PA ) for post-processing the traces. The E P A gives feedback to the mapping and scheduling ( TaskGrapher ) tool by providing actual performance data. These tools are being developed as a part of Parallel Programming Support Environment ( P P S E) research . Results indicate that when actual grain execution time is made available to scheduling tools, accurate projections of program behavior are obtained.
Resource Type
Date Available
Date Issued
Degree Level
Degree Name
Degree Field
Degree Grantor
Commencement Year
Advisor
Academic Affiliation
Non-Academic Affiliation
Subject
Urheberrechts-Erklärung
Publisher
Peer Reviewed
Language
Digitization Specifications
  • File scanned at 300 ppi (Monochrome) using Capture Perfect 3.0.82 on a Canon DR-9080C in PDF format. CVista PdfCompressor 4.0 was used for pdf compression and textual OCR.
Replaces

Beziehungen

Parents:

This work has no parents.

In Collection:

Artikel