Identifying Opportunities for Worst-case Execution Time Reduction in an Avionics System

G. Bernat; R.I. Davis; N. Merriam; J. Tuffen; A. Gardner; M. Bennett; D. Armstrong

Identifying Opportunities for Worst-case Execution Time Reduction in an Avionics System

G. Bernat, R.I. Davis, N. Merriam, J. Tuffen, A. Gardner, M. Bennett, D. Armstrong

Computer Science

Research output: Contribution to journal › Article › peer-review

Abstract

This paper describes the results of a study that identified opportunities for worst-case execution time reduction in the Operational Flight Program (OFP) software of BAE Systems’ Hawk Mission Computer. The RapiTime toolset was used to provide the execution time analysis information required to target optimizations where they would be most effective. Potential optimizations were identified for worst-case hotspots at three levels: design level, sub-program level and low-level. These hotspots accounted for only 1.2% of the lines of code but contributed 29% of the overall execution time. Focused optimizations on these hotspots resulted in a 23% reduction of the overall execution time for the analysed code.

Original language	English
Pages (from-to)	189-194
Journal	Ada User Journal
Volume	28
Issue number	3
Publication status	Published - Sept 2007

Keywords

real-time
WCET analysis
avionics
case study
rapita systems
rapitime

Access to Document

Cite this

@article{b79bab08f67245c3b6b95d7feff7fee8,

title = "Identifying Opportunities for Worst-case Execution Time Reduction in an Avionics System",

abstract = "This paper describes the results of a study that identified opportunities for worst-case execution time reduction in the Operational Flight Program (OFP) software of BAE Systems{\textquoteright} Hawk Mission Computer. The RapiTime toolset was used to provide the execution time analysis information required to target optimizations where they would be most effective. Potential optimizations were identified for worst-case hotspots at three levels: design level, sub-program level and low-level. These hotspots accounted for only 1.2% of the lines of code but contributed 29% of the overall execution time. Focused optimizations on these hotspots resulted in a 23% reduction of the overall execution time for the analysed code.",

keywords = "real-time, WCET analysis, avionics, case study, rapita systems, rapitime",

author = "G. Bernat and R.I. Davis and N. Merriam and J. Tuffen and A. Gardner and M. Bennett and D. Armstrong",

year = "2007",

month = sep,

language = "English",

volume = "28",

pages = "189--194",

journal = "Ada User Journal",

publisher = "Ada-Europe",

number = "3",

}

TY - JOUR

T1 - Identifying Opportunities for Worst-case Execution Time Reduction in an Avionics System

AU - Bernat, G.

AU - Davis, R.I.

AU - Merriam, N.

AU - Tuffen, J.

AU - Gardner, A.

AU - Bennett, M.

AU - Armstrong, D.

PY - 2007/9

Y1 - 2007/9

N2 - This paper describes the results of a study that identified opportunities for worst-case execution time reduction in the Operational Flight Program (OFP) software of BAE Systems’ Hawk Mission Computer. The RapiTime toolset was used to provide the execution time analysis information required to target optimizations where they would be most effective. Potential optimizations were identified for worst-case hotspots at three levels: design level, sub-program level and low-level. These hotspots accounted for only 1.2% of the lines of code but contributed 29% of the overall execution time. Focused optimizations on these hotspots resulted in a 23% reduction of the overall execution time for the analysed code.

AB - This paper describes the results of a study that identified opportunities for worst-case execution time reduction in the Operational Flight Program (OFP) software of BAE Systems’ Hawk Mission Computer. The RapiTime toolset was used to provide the execution time analysis information required to target optimizations where they would be most effective. Potential optimizations were identified for worst-case hotspots at three levels: design level, sub-program level and low-level. These hotspots accounted for only 1.2% of the lines of code but contributed 29% of the overall execution time. Focused optimizations on these hotspots resulted in a 23% reduction of the overall execution time for the analysed code.

KW - real-time

KW - WCET analysis

KW - avionics

KW - case study

KW - rapita systems

KW - rapitime

M3 - Article

VL - 28

SP - 189

EP - 194

JO - Ada User Journal

JF - Ada User Journal

IS - 3

ER -