FPGA-based software profiler for hardware/software co-design

E. Sayed M Saad, Medhat H A Awadalla, Kareem Ezz E-Deen

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

Embedded systems are a mixture of software running on a microprocessor and application-specific hardware . Hardware/Software co-design requires an appropriate profiler to detect the functions that contribute to a large percentage of program execution . Software based profiling tools, such as the well-known GNU gprof profiler, integrates an extra code with the software program to be profiled causing a significant performance overhead. To address this issue, this paper proposes a software profiler called AddressTracer. This profiler is an adaptation of a non-intrusive, real time profiler called SnoopP. the AddressTraeer is accurately able to evaluate the performance matrices of any specific software function. A software benchmark, Secure Hash Algorithm (SHA), is profiled using AddressTracer and other software profiling tools, Airwolf, and GNU software profiling tool (gprof), for a quantitative comparison and their performance overhead are studied. the achieved results show that AddressTracer provides accurate profiling results with no performance overhead. Airwolf causes a very low remarkable performance overhead compared with that incurred by gprof.

Original languageEnglish
Title of host publicationNational Radio Science Conference, NRSC, Proceedings
Publication statusPublished - 2009
Event2009 National Radio Science Conference, NRSC 2009 - New Cairo, Egypt
Duration: Mar 17 2009Mar 19 2009

Other

Other2009 National Radio Science Conference, NRSC 2009
CountryEgypt
CityNew Cairo
Period3/17/093/19/09

Fingerprint

Computer hardware
Field programmable gate arrays (FPGA)
hardware
computer programs
Embedded systems
Microprocessor chips
microprocessors
causes
matrices

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Condensed Matter Physics
  • Electronic, Optical and Magnetic Materials

Cite this

Saad, E. S. M., Awadalla, M. H. A., & E-Deen, K. E. (2009). FPGA-based software profiler for hardware/software co-design. In National Radio Science Conference, NRSC, Proceedings [5233941]

FPGA-based software profiler for hardware/software co-design. / Saad, E. Sayed M; Awadalla, Medhat H A; E-Deen, Kareem Ezz.

National Radio Science Conference, NRSC, Proceedings. 2009. 5233941.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Saad, ESM, Awadalla, MHA & E-Deen, KE 2009, FPGA-based software profiler for hardware/software co-design. in National Radio Science Conference, NRSC, Proceedings., 5233941, 2009 National Radio Science Conference, NRSC 2009, New Cairo, Egypt, 3/17/09.
Saad ESM, Awadalla MHA, E-Deen KE. FPGA-based software profiler for hardware/software co-design. In National Radio Science Conference, NRSC, Proceedings. 2009. 5233941
Saad, E. Sayed M ; Awadalla, Medhat H A ; E-Deen, Kareem Ezz. / FPGA-based software profiler for hardware/software co-design. National Radio Science Conference, NRSC, Proceedings. 2009.
@inproceedings{e875a425441243f5934d77d8518c4e1c,
title = "FPGA-based software profiler for hardware/software co-design",
abstract = "Embedded systems are a mixture of software running on a microprocessor and application-specific hardware . Hardware/Software co-design requires an appropriate profiler to detect the functions that contribute to a large percentage of program execution . Software based profiling tools, such as the well-known GNU gprof profiler, integrates an extra code with the software program to be profiled causing a significant performance overhead. To address this issue, this paper proposes a software profiler called AddressTracer. This profiler is an adaptation of a non-intrusive, real time profiler called SnoopP. the AddressTraeer is accurately able to evaluate the performance matrices of any specific software function. A software benchmark, Secure Hash Algorithm (SHA), is profiled using AddressTracer and other software profiling tools, Airwolf, and GNU software profiling tool (gprof), for a quantitative comparison and their performance overhead are studied. the achieved results show that AddressTracer provides accurate profiling results with no performance overhead. Airwolf causes a very low remarkable performance overhead compared with that incurred by gprof.",
author = "Saad, {E. Sayed M} and Awadalla, {Medhat H A} and E-Deen, {Kareem Ezz}",
year = "2009",
language = "English",
isbn = "9781424442140",
booktitle = "National Radio Science Conference, NRSC, Proceedings",

}

TY - GEN

T1 - FPGA-based software profiler for hardware/software co-design

AU - Saad, E. Sayed M

AU - Awadalla, Medhat H A

AU - E-Deen, Kareem Ezz

PY - 2009

Y1 - 2009

N2 - Embedded systems are a mixture of software running on a microprocessor and application-specific hardware . Hardware/Software co-design requires an appropriate profiler to detect the functions that contribute to a large percentage of program execution . Software based profiling tools, such as the well-known GNU gprof profiler, integrates an extra code with the software program to be profiled causing a significant performance overhead. To address this issue, this paper proposes a software profiler called AddressTracer. This profiler is an adaptation of a non-intrusive, real time profiler called SnoopP. the AddressTraeer is accurately able to evaluate the performance matrices of any specific software function. A software benchmark, Secure Hash Algorithm (SHA), is profiled using AddressTracer and other software profiling tools, Airwolf, and GNU software profiling tool (gprof), for a quantitative comparison and their performance overhead are studied. the achieved results show that AddressTracer provides accurate profiling results with no performance overhead. Airwolf causes a very low remarkable performance overhead compared with that incurred by gprof.

AB - Embedded systems are a mixture of software running on a microprocessor and application-specific hardware . Hardware/Software co-design requires an appropriate profiler to detect the functions that contribute to a large percentage of program execution . Software based profiling tools, such as the well-known GNU gprof profiler, integrates an extra code with the software program to be profiled causing a significant performance overhead. To address this issue, this paper proposes a software profiler called AddressTracer. This profiler is an adaptation of a non-intrusive, real time profiler called SnoopP. the AddressTraeer is accurately able to evaluate the performance matrices of any specific software function. A software benchmark, Secure Hash Algorithm (SHA), is profiled using AddressTracer and other software profiling tools, Airwolf, and GNU software profiling tool (gprof), for a quantitative comparison and their performance overhead are studied. the achieved results show that AddressTracer provides accurate profiling results with no performance overhead. Airwolf causes a very low remarkable performance overhead compared with that incurred by gprof.

UR - http://www.scopus.com/inward/record.url?scp=70449514099&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70449514099&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9781424442140

BT - National Radio Science Conference, NRSC, Proceedings

ER -