Client layer becomes bottleneck: workload analysis of an ultra-large-scale cloud storage system

Proceedings of the 14th IEEE/ACM International Conference on Utility and Cloud Computing Companion Pub Date : 2021-12-06 DOI:10.1145/3492323.3495625

Xiaoyi Sun, Kai Li, Yaodanjun Ren, Jiale Lin, Zhenyu Ren, Shuzhi Feng, Jian Yin, Zhengwei Qi

{"title":"Client layer becomes bottleneck: workload analysis of an ultra-large-scale cloud storage system","authors":"Xiaoyi Sun, Kai Li, Yaodanjun Ren, Jiale Lin, Zhenyu Ren, Shuzhi Feng, Jian Yin, Zhengwei Qi","doi":"10.1145/3492323.3495625","DOIUrl":null,"url":null,"abstract":"Recent years have witnessed the fast development of file and storage systems. Many improvements of file and storage systems are inspired by Workload analysis, which reveals the characteristics of I/O behavior. Although cloud storage systems are becoming increasingly prominent, few real-world and large-scale cloud storage workload studies are presented. Alibaba Cloud is one of the world's largest cloud providers, and we have collected and analyzed workloads from Alibaba for an extended period. We observe that modern cloud network architecture can easily handle the peak load during busy festivals. However, the client layer is the system bottleneck during the peak period, which calls for further optimization. We also find that the workload is heavily skewed toward a small percentage of virtual disks, and its distribution conforms 80/20 rule. In summary, the characteristics of such a large-scale cloud storage system in production environments are important for future cloud storage system modifications.","PeriodicalId":440884,"journal":{"name":"Proceedings of the 14th IEEE/ACM International Conference on Utility and Cloud Computing Companion","volume":"47 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2021-12-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the 14th IEEE/ACM International Conference on Utility and Cloud Computing Companion","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/3492323.3495625","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 0

Abstract

Recent years have witnessed the fast development of file and storage systems. Many improvements of file and storage systems are inspired by Workload analysis, which reveals the characteristics of I/O behavior. Although cloud storage systems are becoming increasingly prominent, few real-world and large-scale cloud storage workload studies are presented. Alibaba Cloud is one of the world's largest cloud providers, and we have collected and analyzed workloads from Alibaba for an extended period. We observe that modern cloud network architecture can easily handle the peak load during busy festivals. However, the client layer is the system bottleneck during the peak period, which calls for further optimization. We also find that the workload is heavily skewed toward a small percentage of virtual disks, and its distribution conforms 80/20 rule. In summary, the characteristics of such a large-scale cloud storage system in production environments are important for future cloud storage system modifications.

查看原文

微信好友朋友圈 QQ好友复制链接

本刊更多论文

客户端层成为瓶颈:超大规模云存储系统的工作负载分析

近年来，文件和存储系统得到了快速发展。工作负载分析启发了文件和存储系统的许多改进，它揭示了I/O行为的特征。虽然云存储系统变得越来越突出，但很少有现实世界和大规模云存储工作负载的研究。阿里云是全球最大的云提供商之一，我们长期收集和分析来自阿里的工作负载。我们观察到，现代云网络架构可以轻松处理繁忙节日期间的峰值负载。但是，客户端层是高峰期的系统瓶颈，需要进一步优化。我们还发现，工作负载严重偏向一小部分虚拟磁盘，其分布符合80/20规则。综上所述，这种大规模云存储系统在生产环境中的特点对未来云存储系统的修改非常重要。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文去求助

来源期刊

Proceedings of the 14th IEEE/ACM International Conference on Utility and Cloud Computing Companion

自引率

0.00%

发文量

期刊最新文献

Blockchain-based distributed platform for accountable medical data sharing An empirical analysis of LADA diabetes case, control and variable importance Estimating the capacities of function-as-a-service functions Session details: International Workshop on Machine Learning and Health Informatics (MLHI) Alcoholism detection via GLCM and particle swarm optimization