Parametric and Non-Parametric Statistics for Program Performance Analysis and Comparison - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Rapport (Rapport De Recherche) Année : 2016

Parametric and Non-Parametric Statistics for Program Performance Analysis and Comparison

Résumé

This report is a continuation of our previous research effort on statistical program performance analysis and comparison, in presence of program performance variability. In the previous study, we gave a formal statistical methodology to analyse program speedups based on mean or median performance metrics: execution time, energy consumption, etc. However mean or median observed performances do not always reflect the user's feeling of performance, especially when the performances are really instable. In the current study, we propose additional precise performance metrics, based on performance modelling using gaussian mixtures. We explore the difference between parametric and non parametric statistics applied on program performance analysis. Our additional statistical metrics for analysing and comparing program performances give the user more precise decision tools to select best code versions, not necessarily based on mean or median numbers. Also, we provide a new metric to estimate performance variability based on gaussian mixture model. Our statistical methods are implemented with R and distributed as open source code.
Fichier principal
Vignette du fichier
RR-8875.pdf (1.69 Mo) Télécharger le fichier
VARCORE_code.zip (22.64 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01286112 , version 1 (14-03-2016)
hal-01286112 , version 2 (23-11-2016)
hal-01286112 , version 3 (29-06-2017)

Licence

Domaine public

Identifiants

  • HAL Id : hal-01286112 , version 2

Citer

Julien Worms, Sid Touati. Parametric and Non-Parametric Statistics for Program Performance Analysis and Comparison. [Research Report] RR-8875, INRIA Sophia Antipolis - I3S; Université Nice Sophia Antipolis; Université Versailles Saint Quentin en Yvelines; Laboratoire de mathématiques de Versailles. 2016, pp.70. ⟨hal-01286112v2⟩
1109 Consultations
4160 Téléchargements

Partager

Gmail Facebook X LinkedIn More