10Gb/s接口虚拟机在Openstack热迁移中的实际评估

M. Biswas, G. Parr, S. McClean, P. Morrow, B. Scotney
{"title":"10Gb/s接口虚拟机在Openstack热迁移中的实际评估","authors":"M. Biswas, G. Parr, S. McClean, P. Morrow, B. Scotney","doi":"10.1109/SOSE.2016.35","DOIUrl":null,"url":null,"abstract":"Live Migration (LM) of Virtual Machine (VM) is a process of transferring a working VM from on host to another host of a different physical machine without interfering the VM. In datacentre networks, LM enables flexibility in resource optimisation, fault tolerance and load balancing. However, in real time, the resource consumption and latency of live VM migration reduce these benefits to much less than their potential. In this paper, we present the results of an experimental study that evaluates LM in our unique high speed optical fibre network connecting Northern Ireland, Dublin and Halifax (Canada). We observe that using Pre-Copy LM extra large amounts of stressed memory leads to non convergence over high latency paths. However, using Post-copy LM the total migration time as well as downtime is dominated by specific memory utilisation patterns inside the virtualised guest. We experience variation in total migration time and downtime using Post-Copy LM considering Quality of Service (QoS) parameters, which can have significant impact in the cloud applications performance.","PeriodicalId":153118,"journal":{"name":"2016 IEEE Symposium on Service-Oriented System Engineering (SOSE)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2016-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"14","resultStr":"{\"title\":\"A Practical Evaluation in Openstack Live Migration of VMs Using 10Gb/s Interfaces\",\"authors\":\"M. Biswas, G. Parr, S. McClean, P. Morrow, B. Scotney\",\"doi\":\"10.1109/SOSE.2016.35\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Live Migration (LM) of Virtual Machine (VM) is a process of transferring a working VM from on host to another host of a different physical machine without interfering the VM. In datacentre networks, LM enables flexibility in resource optimisation, fault tolerance and load balancing. However, in real time, the resource consumption and latency of live VM migration reduce these benefits to much less than their potential. In this paper, we present the results of an experimental study that evaluates LM in our unique high speed optical fibre network connecting Northern Ireland, Dublin and Halifax (Canada). We observe that using Pre-Copy LM extra large amounts of stressed memory leads to non convergence over high latency paths. However, using Post-copy LM the total migration time as well as downtime is dominated by specific memory utilisation patterns inside the virtualised guest. We experience variation in total migration time and downtime using Post-Copy LM considering Quality of Service (QoS) parameters, which can have significant impact in the cloud applications performance.\",\"PeriodicalId\":153118,\"journal\":{\"name\":\"2016 IEEE Symposium on Service-Oriented System Engineering (SOSE)\",\"volume\":\"1 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2016-05-19\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"14\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2016 IEEE Symposium on Service-Oriented System Engineering (SOSE)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/SOSE.2016.35\",\"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 Symposium on Service-Oriented System Engineering (SOSE)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/SOSE.2016.35","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 14

摘要

虚拟机的实时迁移(LM)是在不干扰虚拟机的情况下,将正在工作的虚拟机从一台主机转移到另一台物理机的另一台主机的过程。在数据中心网络中,LM可以实现资源优化、容错和负载平衡方面的灵活性。然而,在实时情况下,虚拟机迁移的资源消耗和延迟会大大降低这些好处。在本文中,我们提出了一项实验研究的结果,该研究评估了我们连接北爱尔兰,都柏林和哈利法克斯(加拿大)的独特高速光纤网络中的LM。我们观察到,使用Pre-Copy LM的超大数量的压力内存会导致高延迟路径上的不收敛。但是,使用后复制LM,总迁移时间和停机时间由虚拟客户机内部的特定内存利用模式决定。在使用Post-Copy LM时,考虑到服务质量(QoS)参数,我们经历了总迁移时间和停机时间的变化,这可能对云应用程序性能产生重大影响。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
A Practical Evaluation in Openstack Live Migration of VMs Using 10Gb/s Interfaces
Live Migration (LM) of Virtual Machine (VM) is a process of transferring a working VM from on host to another host of a different physical machine without interfering the VM. In datacentre networks, LM enables flexibility in resource optimisation, fault tolerance and load balancing. However, in real time, the resource consumption and latency of live VM migration reduce these benefits to much less than their potential. In this paper, we present the results of an experimental study that evaluates LM in our unique high speed optical fibre network connecting Northern Ireland, Dublin and Halifax (Canada). We observe that using Pre-Copy LM extra large amounts of stressed memory leads to non convergence over high latency paths. However, using Post-copy LM the total migration time as well as downtime is dominated by specific memory utilisation patterns inside the virtualised guest. We experience variation in total migration time and downtime using Post-Copy LM considering Quality of Service (QoS) parameters, which can have significant impact in the cloud applications performance.
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:481959085
Book学术官方微信