定时器中断间隔对文件系统同步开销的影响

H. Son, Seongjin Lee, Y. Won
{"title":"定时器中断间隔对文件系统同步开销的影响","authors":"H. Son, Seongjin Lee, Y. Won","doi":"10.1109/ICNIDC.2016.7974543","DOIUrl":null,"url":null,"abstract":"File system metadata is indispensable in both describing the data and maintaining the file system. Despite the importance of metadata in the file system, the overhead of maintaining the metadata cannot be taken lightly. It is because the metadata also have to be persisted on the storage device and it consumes IO bandwidth as well as creates journaling overhead. In this paper, we find that the random write with synchronous performance of a storage is significantly affected by not only the hardware performance but also timer interrupt interval of the kernel. Extending the timer interrupt interval allows reducing the write volume and increasing the random write followed by fsync() performance of EXT4 file system. We propose intermittent mtime timestamp update on Coarse grain mtime interval instead of Fine grain mtime interval. The experiment results with mtime update interval of 1 second show that the total write volume is decreased by 75% and 28%, respectively compared to total write volume of 1 ms and 10 ms mtime interval, and the throughput increased 3.1× and 1.2× compared to 1 ms and 10 ms mtime interval. Coarse grain mtime update is resolve to the journaling overhead issues while still logging mtime timestamp.","PeriodicalId":439987,"journal":{"name":"2016 IEEE International Conference on Network Infrastructure and Digital Content (IC-NIDC)","volume":"70 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2016-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":"{\"title\":\"Effect of timer interrupt interval on file system synchronization overhead\",\"authors\":\"H. Son, Seongjin Lee, Y. Won\",\"doi\":\"10.1109/ICNIDC.2016.7974543\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"File system metadata is indispensable in both describing the data and maintaining the file system. Despite the importance of metadata in the file system, the overhead of maintaining the metadata cannot be taken lightly. It is because the metadata also have to be persisted on the storage device and it consumes IO bandwidth as well as creates journaling overhead. In this paper, we find that the random write with synchronous performance of a storage is significantly affected by not only the hardware performance but also timer interrupt interval of the kernel. Extending the timer interrupt interval allows reducing the write volume and increasing the random write followed by fsync() performance of EXT4 file system. We propose intermittent mtime timestamp update on Coarse grain mtime interval instead of Fine grain mtime interval. The experiment results with mtime update interval of 1 second show that the total write volume is decreased by 75% and 28%, respectively compared to total write volume of 1 ms and 10 ms mtime interval, and the throughput increased 3.1× and 1.2× compared to 1 ms and 10 ms mtime interval. Coarse grain mtime update is resolve to the journaling overhead issues while still logging mtime timestamp.\",\"PeriodicalId\":439987,\"journal\":{\"name\":\"2016 IEEE International Conference on Network Infrastructure and Digital Content (IC-NIDC)\",\"volume\":\"70 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2016-09-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"1\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2016 IEEE International Conference on Network Infrastructure and Digital Content (IC-NIDC)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ICNIDC.2016.7974543\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2016 IEEE International Conference on Network Infrastructure and Digital Content (IC-NIDC)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICNIDC.2016.7974543","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1

摘要

文件系统元数据在描述数据和维护文件系统方面都是不可或缺的。尽管元数据在文件系统中很重要,但维护元数据的开销不能掉以轻心。这是因为元数据还必须持久化在存储设备上,并且它消耗IO带宽并创建日志开销。在本文中,我们发现随机写入的同步性能不仅受到硬件性能的显著影响,而且受到内核定时器中断间隔的显著影响。延长定时器中断间隔可以减少EXT4文件系统的写入量,增加随机写入,从而提高fsync()性能。我们提出在粗粒度mtime间隔上进行间歇mtime时间戳更新,而不是在细粒度mtime间隔上。实验结果表明,mtime更新间隔为1秒时,与mtime间隔为1 ms和10 ms时相比,总写容量分别减少了75%和28%,吞吐量比mtime间隔为1 ms和10 ms时分别提高了3.1倍和1.2倍。粗粒度mtime更新解决了日志开销问题,同时仍然记录mtime时间戳。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Effect of timer interrupt interval on file system synchronization overhead
File system metadata is indispensable in both describing the data and maintaining the file system. Despite the importance of metadata in the file system, the overhead of maintaining the metadata cannot be taken lightly. It is because the metadata also have to be persisted on the storage device and it consumes IO bandwidth as well as creates journaling overhead. In this paper, we find that the random write with synchronous performance of a storage is significantly affected by not only the hardware performance but also timer interrupt interval of the kernel. Extending the timer interrupt interval allows reducing the write volume and increasing the random write followed by fsync() performance of EXT4 file system. We propose intermittent mtime timestamp update on Coarse grain mtime interval instead of Fine grain mtime interval. The experiment results with mtime update interval of 1 second show that the total write volume is decreased by 75% and 28%, respectively compared to total write volume of 1 ms and 10 ms mtime interval, and the throughput increased 3.1× and 1.2× compared to 1 ms and 10 ms mtime interval. Coarse grain mtime update is resolve to the journaling overhead issues while still logging mtime timestamp.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信