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
L'enregistrement/relecture est un outil essentiel dans les cloud pour fournir de nombreuses fonctionnalités telles que la tolérance aux pannes, le débogage logiciel et l'analyse de sécurité en enregistrant l'exécution dans un journal et en la rejouant de manière déterministe ultérieurement. Cependant, dans les environnements virtualisés, le fichier journal augmente considérablement en raison de la sauvegarde d'une quantité considérable de données d'E/S, introduisant finalement des coûts de stockage importants. Pour atténuer ce problème, cet article propose RR-Row, un disque de machine virtuelle basé sur la redirection sur écriture pour les scénarios d'enregistrement/relecture. RR-Row ajoute les données écrites dans de nouveaux blocs plutôt que d'écraser les blocs d'origine lors de l'exécution normale afin que toutes les données écrites soient réservées sur le disque. De cette façon, le système d'enregistrement enregistre uniquement l'identifiant du bloc au lieu du contenu complet, et le système de relecture peut récupérer directement les données du disque plutôt que du journal, réduisant ainsi considérablement la taille du journal. De plus, nous proposons plusieurs optimisations pour améliorer les performances des E/S afin qu'elles soient également adaptées à une exécution normale. Nous implémentons RR-Row pour QEMU et menons un ensemble d'expériences. Les résultats montrent que RR-Row réduit la taille des journaux de 68 % par rapport au disque Raw/QCow2 actuellement utilisé sans compromettre les performances d'E/S.
Ying ZHAO
Guilin University of Electronic Technology
Youquan XIAN
Guangxi Normal University
Yongnan LI
People's Public Security University of China
Peng LIU
Guangxi Normal University
Dongcheng LI
Guangxi Normal University
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
Ying ZHAO, Youquan XIAN, Yongnan LI, Peng LIU, Dongcheng LI, "RR-Row: Redirect-on-Write Based Virtual Machine Disk for Record/Replay" in IEICE TRANSACTIONS on Information,
vol. E107-D, no. 2, pp. 169-179, February 2024, doi: 10.1587/transinf.2023EDP7122.
Abstract: Record/replay is one essential tool in clouds to provide many capabilities such as fault tolerance, software debugging, and security analysis by recording the execution into a log and replaying it deterministically later on. However, in virtualized environments, the log file increases heavily due to saving a considerable amount of I/O data, finally introducing significant storage costs. To mitigate this problem, this paper proposes RR-Row, a redirect-on-write based virtual machine disk for record/replay scenarios. RR-Row appends the written data into new blocks rather than overwrites the original blocks during normal execution so that all written data are reserved in the disk. In this way, the record system only saves the block id instead of the full content, and the replay system can directly fetch the data from the disk rather than the log, thereby reducing the log size a lot. In addition, we propose several optimizations for improving I/O performance so that it is also suitable for normal execution. We implement RR-Row for QEMU and conduct a set of experiments. The results show that RR-Row reduces the log size by 68% compared to the currently used Raw/QCow2 disk without compromising I/O performance.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2023EDP7122/_p
Copier
@ARTICLE{e107-d_2_169,
author={Ying ZHAO, Youquan XIAN, Yongnan LI, Peng LIU, Dongcheng LI, },
journal={IEICE TRANSACTIONS on Information},
title={RR-Row: Redirect-on-Write Based Virtual Machine Disk for Record/Replay},
year={2024},
volume={E107-D},
number={2},
pages={169-179},
abstract={Record/replay is one essential tool in clouds to provide many capabilities such as fault tolerance, software debugging, and security analysis by recording the execution into a log and replaying it deterministically later on. However, in virtualized environments, the log file increases heavily due to saving a considerable amount of I/O data, finally introducing significant storage costs. To mitigate this problem, this paper proposes RR-Row, a redirect-on-write based virtual machine disk for record/replay scenarios. RR-Row appends the written data into new blocks rather than overwrites the original blocks during normal execution so that all written data are reserved in the disk. In this way, the record system only saves the block id instead of the full content, and the replay system can directly fetch the data from the disk rather than the log, thereby reducing the log size a lot. In addition, we propose several optimizations for improving I/O performance so that it is also suitable for normal execution. We implement RR-Row for QEMU and conduct a set of experiments. The results show that RR-Row reduces the log size by 68% compared to the currently used Raw/QCow2 disk without compromising I/O performance.},
keywords={},
doi={10.1587/transinf.2023EDP7122},
ISSN={1745-1361},
month={February},}
Copier
TY - JOUR
TI - RR-Row: Redirect-on-Write Based Virtual Machine Disk for Record/Replay
T2 - IEICE TRANSACTIONS on Information
SP - 169
EP - 179
AU - Ying ZHAO
AU - Youquan XIAN
AU - Yongnan LI
AU - Peng LIU
AU - Dongcheng LI
PY - 2024
DO - 10.1587/transinf.2023EDP7122
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E107-D
IS - 2
JA - IEICE TRANSACTIONS on Information
Y1 - February 2024
AB - Record/replay is one essential tool in clouds to provide many capabilities such as fault tolerance, software debugging, and security analysis by recording the execution into a log and replaying it deterministically later on. However, in virtualized environments, the log file increases heavily due to saving a considerable amount of I/O data, finally introducing significant storage costs. To mitigate this problem, this paper proposes RR-Row, a redirect-on-write based virtual machine disk for record/replay scenarios. RR-Row appends the written data into new blocks rather than overwrites the original blocks during normal execution so that all written data are reserved in the disk. In this way, the record system only saves the block id instead of the full content, and the replay system can directly fetch the data from the disk rather than the log, thereby reducing the log size a lot. In addition, we propose several optimizations for improving I/O performance so that it is also suitable for normal execution. We implement RR-Row for QEMU and conduct a set of experiments. The results show that RR-Row reduces the log size by 68% compared to the currently used Raw/QCow2 disk without compromising I/O performance.
ER -