Anthony K. Tsetse, Samuel Tweneboah-Koduah, B. Rawal, Zhihao Zheng, Manoah Prattipati
{"title":"系统虚拟化性能的比较研究","authors":"Anthony K. Tsetse, Samuel Tweneboah-Koduah, B. Rawal, Zhihao Zheng, Manoah Prattipati","doi":"10.1109/IRI.2019.00064","DOIUrl":null,"url":null,"abstract":"Virtualization presents an abstraction between the bare computer hardware (physical resource set) and the application running on top of it. Virtualization technology has become the defacto technology in cloud computing systems. The technique has become an enabling technology in cloud computing, which helps in the provisioning of resources on an on-demand basis, thereby addressing elastic resource requirements of organizations in their quest to scale. This paper seeks to investigate the effect of virtualization on system performance. The primary objective is to conduct an empirical analysis to compare and contrast the performance of Virtual Machines (VMs) and a Host machine running on different Operating Systems (OS). The effect of the use of different OS and VM configurations (Host and Guest systems) on system performance is also explored. The study shows that depending on the metric (i.e., throughput, jitter, and response time and packet loss ratio) used, virtualization technique results in a performance degradation between 21% and 39%. In particular, communication between systems with Linux (being either the Host or the Guest OS) performs better than Windows (as either the Guest or the Host) OS. Additionally, it is also observed that there is a performance loss ranging between 12% and 20% when communicating devices run heterogeneous Guest OS.","PeriodicalId":295028,"journal":{"name":"2019 IEEE 20th International Conference on Information Reuse and Integration for Data Science (IRI)","volume":"89 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"A Comparative Study of System Virtualization Performance\",\"authors\":\"Anthony K. Tsetse, Samuel Tweneboah-Koduah, B. Rawal, Zhihao Zheng, Manoah Prattipati\",\"doi\":\"10.1109/IRI.2019.00064\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Virtualization presents an abstraction between the bare computer hardware (physical resource set) and the application running on top of it. Virtualization technology has become the defacto technology in cloud computing systems. The technique has become an enabling technology in cloud computing, which helps in the provisioning of resources on an on-demand basis, thereby addressing elastic resource requirements of organizations in their quest to scale. This paper seeks to investigate the effect of virtualization on system performance. The primary objective is to conduct an empirical analysis to compare and contrast the performance of Virtual Machines (VMs) and a Host machine running on different Operating Systems (OS). The effect of the use of different OS and VM configurations (Host and Guest systems) on system performance is also explored. The study shows that depending on the metric (i.e., throughput, jitter, and response time and packet loss ratio) used, virtualization technique results in a performance degradation between 21% and 39%. In particular, communication between systems with Linux (being either the Host or the Guest OS) performs better than Windows (as either the Guest or the Host) OS. Additionally, it is also observed that there is a performance loss ranging between 12% and 20% when communicating devices run heterogeneous Guest OS.\",\"PeriodicalId\":295028,\"journal\":{\"name\":\"2019 IEEE 20th International Conference on Information Reuse and Integration for Data Science (IRI)\",\"volume\":\"89 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2019-07-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2019 IEEE 20th International Conference on Information Reuse and Integration for Data Science (IRI)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/IRI.2019.00064\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2019 IEEE 20th International Conference on Information Reuse and Integration for Data Science (IRI)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/IRI.2019.00064","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
A Comparative Study of System Virtualization Performance
Virtualization presents an abstraction between the bare computer hardware (physical resource set) and the application running on top of it. Virtualization technology has become the defacto technology in cloud computing systems. The technique has become an enabling technology in cloud computing, which helps in the provisioning of resources on an on-demand basis, thereby addressing elastic resource requirements of organizations in their quest to scale. This paper seeks to investigate the effect of virtualization on system performance. The primary objective is to conduct an empirical analysis to compare and contrast the performance of Virtual Machines (VMs) and a Host machine running on different Operating Systems (OS). The effect of the use of different OS and VM configurations (Host and Guest systems) on system performance is also explored. The study shows that depending on the metric (i.e., throughput, jitter, and response time and packet loss ratio) used, virtualization technique results in a performance degradation between 21% and 39%. In particular, communication between systems with Linux (being either the Host or the Guest OS) performs better than Windows (as either the Guest or the Host) OS. Additionally, it is also observed that there is a performance loss ranging between 12% and 20% when communicating devices run heterogeneous Guest OS.