The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
En raison du développement rapide de différents processeurs, par exemple x86 et Sunway, le portage de logiciels entre différentes plates-formes devient de plus en plus fréquent. Cependant, l'efficacité d'exécution du logiciel migré sur la plate-forme cible est différente de celle de la plate-forme source, et la plupart des études précédentes ont étudié l'amélioration de l'efficacité du point de vue matériel. À notre connaissance, il s'agit du premier article à se concentrer exclusivement sur l'étude des facteurs logiciels qui peuvent entraîner une modification des performances après la migration d'un logiciel. Pour réaliser notre étude, nous avons utilisé SonarQube pour détecter et mesurer cinq facteurs logiciels, à savoir les lignes dupliquées (DL), la densité des odeurs de code (CSD), les grandes fonctions (BF), la complexité cyclomatique (CC) et les fonctions complexes (CF), de 13 projets sélectionnés de la suite de benchmarks SPEC CPU2006. Ensuite, nous avons mesuré l'évolution des performances logicielles en calculant le taux d'accélération du temps d'exécution avant (x86) et après la migration logicielle (Sunway). Enfin, nous avons réalisé un modèle de régression linéaire multiple pour analyser la relation entre le changement de performances du logiciel et les facteurs logiciels. Les résultats indiquent que le changement de performances de la migration logicielle de la plate-forme x86 vers la plate-forme Sunway est principalement affecté par trois facteurs logiciels, à savoir la densité d'odeur de code (CSD), la complexité cyclomatique (CC) et les fonctions complexes (CF). Les résultats peuvent profiter à la fois aux chercheurs et aux praticiens.
Jie TAN
State Key Laboratory of Mathematical Engineering and Advanced Computing
Jianmin PANG
State Key Laboratory of Mathematical Engineering and Advanced Computing
Cong LIU
Shandong University of Technology
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copier
Jie TAN, Jianmin PANG, Cong LIU, "What Factors Affect the Performance of Software after Migration: A Case Study on Sunway TaihuLight Supercomputer" in IEICE TRANSACTIONS on Information,
vol. E105-D, no. 1, pp. 26-30, January 2022, doi: 10.1587/transinf.2021MPL0003.
Abstract: Due to the rapid development of different processors, e.g., x86 and Sunway, software porting between different platforms is becoming more frequent. However, the migrated software's execution efficiency on the target platform is different from that of the source platform, and most of the previous studies have investigated the improvement of the efficiency from the hardware perspective. To the best of our knowledge, this is the first paper to exclusively focus on studying what software factors can result in performance change after software migration. To perform our study, we used SonarQube to detect and measure five software factors, namely Duplicated Lines (DL), Code Smells Density (CSD), Big Functions (BF), Cyclomatic Complexity (CC), and Complex Functions (CF), from 13 selected projects of SPEC CPU2006 benchmark suite. Then, we measured the change of software performance by calculating the acceleration ratio of execution time before (x86) and after (Sunway) software migration. Finally, we performed a multiple linear regression model to analyze the relationship between the software performance change and the software factors. The results indicate that the performance change of software migration from the x86 platform to the Sunway platform is mainly affected by three software factors, i.e., Code Smell Density (CSD), Cyclomatic Complexity (CC), and Complex Functions (CF). The findings can benefit both researchers and practitioners.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2021MPL0003/_p
Copier
@ARTICLE{e105-d_1_26,
author={Jie TAN, Jianmin PANG, Cong LIU, },
journal={IEICE TRANSACTIONS on Information},
title={What Factors Affect the Performance of Software after Migration: A Case Study on Sunway TaihuLight Supercomputer},
year={2022},
volume={E105-D},
number={1},
pages={26-30},
abstract={Due to the rapid development of different processors, e.g., x86 and Sunway, software porting between different platforms is becoming more frequent. However, the migrated software's execution efficiency on the target platform is different from that of the source platform, and most of the previous studies have investigated the improvement of the efficiency from the hardware perspective. To the best of our knowledge, this is the first paper to exclusively focus on studying what software factors can result in performance change after software migration. To perform our study, we used SonarQube to detect and measure five software factors, namely Duplicated Lines (DL), Code Smells Density (CSD), Big Functions (BF), Cyclomatic Complexity (CC), and Complex Functions (CF), from 13 selected projects of SPEC CPU2006 benchmark suite. Then, we measured the change of software performance by calculating the acceleration ratio of execution time before (x86) and after (Sunway) software migration. Finally, we performed a multiple linear regression model to analyze the relationship between the software performance change and the software factors. The results indicate that the performance change of software migration from the x86 platform to the Sunway platform is mainly affected by three software factors, i.e., Code Smell Density (CSD), Cyclomatic Complexity (CC), and Complex Functions (CF). The findings can benefit both researchers and practitioners.},
keywords={},
doi={10.1587/transinf.2021MPL0003},
ISSN={1745-1361},
month={January},}
Copier
TY - JOUR
TI - What Factors Affect the Performance of Software after Migration: A Case Study on Sunway TaihuLight Supercomputer
T2 - IEICE TRANSACTIONS on Information
SP - 26
EP - 30
AU - Jie TAN
AU - Jianmin PANG
AU - Cong LIU
PY - 2022
DO - 10.1587/transinf.2021MPL0003
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E105-D
IS - 1
JA - IEICE TRANSACTIONS on Information
Y1 - January 2022
AB - Due to the rapid development of different processors, e.g., x86 and Sunway, software porting between different platforms is becoming more frequent. However, the migrated software's execution efficiency on the target platform is different from that of the source platform, and most of the previous studies have investigated the improvement of the efficiency from the hardware perspective. To the best of our knowledge, this is the first paper to exclusively focus on studying what software factors can result in performance change after software migration. To perform our study, we used SonarQube to detect and measure five software factors, namely Duplicated Lines (DL), Code Smells Density (CSD), Big Functions (BF), Cyclomatic Complexity (CC), and Complex Functions (CF), from 13 selected projects of SPEC CPU2006 benchmark suite. Then, we measured the change of software performance by calculating the acceleration ratio of execution time before (x86) and after (Sunway) software migration. Finally, we performed a multiple linear regression model to analyze the relationship between the software performance change and the software factors. The results indicate that the performance change of software migration from the x86 platform to the Sunway platform is mainly affected by three software factors, i.e., Code Smell Density (CSD), Cyclomatic Complexity (CC), and Complex Functions (CF). The findings can benefit both researchers and practitioners.
ER -