Going beyond mean and median programs performancesReportar como inadecuado

Going beyond mean and median programs performances - Descarga este documento en PDF. Documentación en PDF para descargar gratis. Disponible también para leer online.

1 LMV - Laboratoire de Mathématiques de Versailles 2 AOSTE - Models and methods of analysis and optimization for systems with real-time and embedding constraints CRISAM - Inria Sophia Antipolis - Méditerranée , COMRED - COMmunications, Réseaux, systèmes Embarqués et Distribués, Inria de Paris 3 I3S - Laboratoire d-Informatique, Signaux, et Systèmes de Sophia Antipolis

Abstract : In the area of code performance optimisation and tuning, we are faced on the difficult problem of selecting the - best - code version based on empirical experiments and statistical analysis. With the massive introduction of general purpose multicore processors, programs performances become more and more instable, especially parallel programs. Usual statistical methods for computing performance speedups and comparing between programs are based on testing mean or median values. In this article, we explain why these metrics may be inadequate for making relevent decisions, and we propose new performance metrics based on parametric statistics using gaussian mixture models. Our new statistical methods are more accurate for decision making, they are formally defined, computed, implemented and distributed as free software in 1.

Keywords : programs performances modelling gaussians mixture programs performances variability parametric statistics

Autor: Julien Worms - Sid Touati -

Fuente: https://hal.archives-ouvertes.fr/


Documentos relacionados