Skip to Main Content
Frequently Asked Questions
Submit an ETD
Global Search Box
Need Help?
Keyword Search
Participating Institutions
Advanced Search
School Logo
Files
File List
ucin1061553201.pdf (783.31 KB)
ETD Abstract Container
Abstract Header
Software Performance Estimation Techniques in a Co-Design Environment
Author Info
Subramanian, Sriram
Permalink:
http://rave.ohiolink.edu/etdc/view?acc_num=ucin1061553201
Abstract Details
Year and Degree
2003, MS, University of Cincinnati, Engineering : Computer Engineering.
Abstract
Cadence’s Virtual Component Co-design (VCC), first of its kind, provides a frame work for hardware-software co-design exploration. Since design trade-offs largely depend on the ability to accurately and quickly estimate the design constraint of interest, better estimation techniques are always sought after. Software estimation techniques vary from fastest, but least accurate (Static Estimation) to slowest, but most accurate (Cosimulation). We investigated the applicability of two most important techniques, namely Virtual Processor Model (VPM) (also known as annotated_c technique) and Cosimulation with an Instruction Set Simulator (CSISS) to VCC’s environment. We focused on evaluating CSISS against performance and time taken. Representative benchmark routines were chosen based on the execution profiles of the MiBench benchmarks and implemented in VCC specific C language called White Box C. Results were compared with timing estimates obtained using MIPS R10000 perfex profiling counters. We observed that the CSISS based estimation technique is more accurate with an average absolute error of 19.38% and the annotated_c technique had an average absolute error of 58.3%. The CSISS technique was slower by 2.56 times on an average.
Committee
Dr. Karen Tomko (Advisor)
Pages
56 p.
Subject Headings
Computer Science
Keywords
software performance estimation
;
VCC
;
SPIM
;
MiBench
;
hardware-software-codesign
Recommended Citations
Refworks
EndNote
RIS
Mendeley
Citations
Subramanian, S. (2003).
Software Performance Estimation Techniques in a Co-Design Environment
[Master's thesis, University of Cincinnati]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1061553201
APA Style (7th edition)
Subramanian, Sriram.
Software Performance Estimation Techniques in a Co-Design Environment.
2003. University of Cincinnati, Master's thesis.
OhioLINK Electronic Theses and Dissertations Center
, http://rave.ohiolink.edu/etdc/view?acc_num=ucin1061553201.
MLA Style (8th edition)
Subramanian, Sriram. "Software Performance Estimation Techniques in a Co-Design Environment." Master's thesis, University of Cincinnati, 2003. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1061553201
Chicago Manual of Style (17th edition)
Abstract Footer
Document number:
ucin1061553201
Download Count:
631
Copyright Info
© 2003, all rights reserved.
This open access ETD is published by University of Cincinnati and OhioLINK.