A Data Analysis Method for Software Performance Prediction

Authors:
G. Bontempi;W. Kruijtzer
Affiliations:
IMEC, Leuven, Belgium;Philips Research Laboratories, Eindhoven, The Netherlands
Venue:
Proceedings of the conference on Design, automation and test in Europe
Year:
2002

Citing 0
Cited 9

Enabling Multimedia QoS Control with Black-Box Modelling

Soft-Ware 2002 Proceedings of the First International Conference on Computing in an Imperfect World
Accurate software performance estimation using domain classification and neural networks

SBCCI '04 Proceedings of the 17th symposium on Integrated circuits and system design
A Platform-Independent Methodology for Performance Estimation of Multimedia Signal Processing Applications

Journal of VLSI Signal Processing Systems
Fast cycle-approximate instruction set simulation

SCOPES '08 Proceedings of the 11th international workshop on Software & compilers for embedded systems
Applying neural networks to performance estimation of embedded software

Journal of Systems Architecture: the EUROMICRO Journal
SciSim: a software performance estimation framework using source code instrumentation

WOSP '08 Proceedings of the 7th international workshop on Software and performance
Using continuous statistical machine learning to enable high-speed performance prediction in hybrid instruction-/cycle-accurate instruction set simulators

CODES+ISSS '09 Proceedings of the 7th IEEE/ACM international conference on Hardware/software codesign and system synthesis
Performance modeling of embedded applications with zero architectural knowledge

CODES/ISSS '10 Proceedings of the eighth IEEE/ACM/IFIP international conference on Hardware/software codesign and system synthesis
Statistical Performance Modeling in Functional Instruction Set Simulators

ACM Transactions on Embedded Computing Systems (TECS)

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper explores the role of data analysis methodsto support system-level designers in characterising theperformance of embedded applications. In particular, weaddress the performance modelling of software applicationsrunning on an embedded microprocessor. Wepropose a data analysis method, which, on the basis of aparameterisation of the software functionality and thehardware architecture, is able to predict the number ofexecution cycles on an embedded processor. Experimentswith standard computational code (sorting, mathematicalcomputation) and with MPEG variable length decodingare presented to support this claim.