SDVC:用于云中的虚拟机映像的可扩展重复数据删除集群

2014 9th IEEE International Conference on Networking, Architecture, and Storage Pub Date : 2014-08-06 DOI:10.1109/NAS.2014.20

Chuan Lin, Q. Cao, Hongliang Zhang, Guo-Hao Huang, C. Xie

{"title":"SDVC:用于云中的虚拟机映像的可扩展重复数据删除集群","authors":"Chuan Lin, Q. Cao, Hongliang Zhang, Guo-Hao Huang, C. Xie","doi":"10.1109/NAS.2014.20","DOIUrl":null,"url":null,"abstract":"Nowadays, while the storage requirement of virtual machine images generated in cloud infrastructures can be potentially reduced by the deduplication, considering their scale and intensity, the deduplication cluster is demanded. Therefore, in this paper we present SDVC, a scalable deduplication cluster for virtual machine images in cloud. SDVC offers both vertical and horizontal scalability. The horizontal scalability is supported by a three-party distributed infrastructure and a hash allocation algorithm. Meanwhile, categorized chunk tracer and buffer capture hot data. Furthermore, SDVC is vertical scalable by setting a suitable hot chunk buffer in virtual machine servers according to their resource usage, reducing chunk searching operations and relieving the workloads on dedup servers. Our experimental results based on a small scale cluster show that the deduplication throughput achieves up to 80% increase with the number of Dedup servers. Furthermore, only hundreds of Kbytes of categoried hot chunk buffer can provide almost 100% performance improvement.","PeriodicalId":186621,"journal":{"name":"2014 9th IEEE International Conference on Networking, Architecture, and Storage","volume":"11 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2014-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":"{\"title\":\"SDVC: A Scalable Deduplication Cluster for Virtual Machine Images in Cloud\",\"authors\":\"Chuan Lin, Q. Cao, Hongliang Zhang, Guo-Hao Huang, C. Xie\",\"doi\":\"10.1109/NAS.2014.20\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Nowadays, while the storage requirement of virtual machine images generated in cloud infrastructures can be potentially reduced by the deduplication, considering their scale and intensity, the deduplication cluster is demanded. Therefore, in this paper we present SDVC, a scalable deduplication cluster for virtual machine images in cloud. SDVC offers both vertical and horizontal scalability. The horizontal scalability is supported by a three-party distributed infrastructure and a hash allocation algorithm. Meanwhile, categorized chunk tracer and buffer capture hot data. Furthermore, SDVC is vertical scalable by setting a suitable hot chunk buffer in virtual machine servers according to their resource usage, reducing chunk searching operations and relieving the workloads on dedup servers. Our experimental results based on a small scale cluster show that the deduplication throughput achieves up to 80% increase with the number of Dedup servers. Furthermore, only hundreds of Kbytes of categoried hot chunk buffer can provide almost 100% performance improvement.\",\"PeriodicalId\":186621,\"journal\":{\"name\":\"2014 9th IEEE International Conference on Networking, Architecture, and Storage\",\"volume\":\"11 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2014-08-06\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"1\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2014 9th IEEE International Conference on Networking, Architecture, and Storage\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/NAS.2014.20\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2014 9th IEEE International Conference on Networking, Architecture, and Storage","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/NAS.2014.20","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

如今，虽然通过重复数据删除可以潜在地降低云基础设施中生成的虚拟机映像的存储需求，但考虑到其规模和强度，需要重复数据删除集群。因此，在本文中，我们提出了SDVC，一个可扩展的重复数据删除集群，用于云中的虚拟机映像。SDVC提供垂直和水平的可伸缩性。横向可伸缩性由三方分布式基础设施和散列分配算法支持。同时，分类块跟踪器和缓冲区捕获热数据。此外，通过在虚拟机服务器中根据资源使用情况设置合适的热块缓冲区，SDVC具有垂直可扩展性，减少了块搜索操作，减轻了dedup服务器的工作负载。基于小型集群的实验结果表明，随着Dedup服务器数量的增加，重复数据删除吞吐量可提高80%。此外，仅数百kb的分类热块缓冲区就可以提供几乎100%的性能改进。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

SDVC: A Scalable Deduplication Cluster for Virtual Machine Images in Cloud

Nowadays, while the storage requirement of virtual machine images generated in cloud infrastructures can be potentially reduced by the deduplication, considering their scale and intensity, the deduplication cluster is demanded. Therefore, in this paper we present SDVC, a scalable deduplication cluster for virtual machine images in cloud. SDVC offers both vertical and horizontal scalability. The horizontal scalability is supported by a three-party distributed infrastructure and a hash allocation algorithm. Meanwhile, categorized chunk tracer and buffer capture hot data. Furthermore, SDVC is vertical scalable by setting a suitable hot chunk buffer in virtual machine servers according to their resource usage, reducing chunk searching operations and relieving the workloads on dedup servers. Our experimental results based on a small scale cluster show that the deduplication throughput achieves up to 80% increase with the number of Dedup servers. Furthermore, only hundreds of Kbytes of categoried hot chunk buffer can provide almost 100% performance improvement.

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

2014 9th IEEE International Conference on Networking, Architecture, and Storage

自引率

0.00%

发文量