{"title":"基于arm的sbc虚拟/本机节点网络性能评估(KVM为Hypervisor","authors":"Eric Gamess","doi":"10.1145/3564746.3587015","DOIUrl":null,"url":null,"abstract":"Single-Board Computers (SBCs) are increasingly being used due to their small form factor, reduced energy consumption, versatility, affordability, and increasing computational power. Therefore, they are now used in projects where they were not initially contemplated, such as running a Virtual Machine Manager (VMM). In this work, two ARM-based SBCs were selected (Raspberry Pi 4 Model B and ODROID-N2+) and an empirical evaluation was carried out to evaluate the network performance between two nodes running in the same SBC, or in different SBCs directly connected through WiFi or Ethernet. Although several hypervisors are suitable for these SBCs, Kernel-based Virtual Machine (KVM) was chosen since it seems to be the most active project that is developed for the ARM-based architecture. The metrics reported in this study include the TCP latency, UDP latency, TCP throughput, and HTTP latency. In general, the network performance of the ODROID-N2+ exceeded the Raspberry Pi 4 Model B. However, the latter has an indisputable advantage over the former with a much larger and more active community, making the development and deployment of applications much faster and straightforward. Hence, selecting the suitable SBCs should be done cautiously, considering the required software and additional hardware that the project is planning to connect to the SBCs.","PeriodicalId":322431,"journal":{"name":"Proceedings of the 2023 ACM Southeast Conference","volume":"46 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2023-04-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Network Performance Evaluation Between Virtual/Native Nodes Running on ARM-based SBCs Using KVM as Hypervisor\",\"authors\":\"Eric Gamess\",\"doi\":\"10.1145/3564746.3587015\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Single-Board Computers (SBCs) are increasingly being used due to their small form factor, reduced energy consumption, versatility, affordability, and increasing computational power. Therefore, they are now used in projects where they were not initially contemplated, such as running a Virtual Machine Manager (VMM). In this work, two ARM-based SBCs were selected (Raspberry Pi 4 Model B and ODROID-N2+) and an empirical evaluation was carried out to evaluate the network performance between two nodes running in the same SBC, or in different SBCs directly connected through WiFi or Ethernet. Although several hypervisors are suitable for these SBCs, Kernel-based Virtual Machine (KVM) was chosen since it seems to be the most active project that is developed for the ARM-based architecture. The metrics reported in this study include the TCP latency, UDP latency, TCP throughput, and HTTP latency. In general, the network performance of the ODROID-N2+ exceeded the Raspberry Pi 4 Model B. However, the latter has an indisputable advantage over the former with a much larger and more active community, making the development and deployment of applications much faster and straightforward. Hence, selecting the suitable SBCs should be done cautiously, considering the required software and additional hardware that the project is planning to connect to the SBCs.\",\"PeriodicalId\":322431,\"journal\":{\"name\":\"Proceedings of the 2023 ACM Southeast Conference\",\"volume\":\"46 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2023-04-12\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Proceedings of the 2023 ACM Southeast Conference\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1145/3564746.3587015\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the 2023 ACM Southeast Conference","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/3564746.3587015","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
摘要
单板计算机(sbc)由于其小尺寸、低能耗、多功能性、可负担性和不断提高的计算能力而越来越多地被使用。因此,它们现在被用于最初没有考虑到的项目中,例如运行虚拟机管理器(VMM)。本文选择了两个基于arm的SBC (Raspberry Pi 4 Model B和ODROID-N2+),对运行在同一SBC中的两个节点,以及通过WiFi或以太网直接连接的不同SBC中的两个节点之间的网络性能进行了实证评估。尽管有几个管理程序适合这些sbc,但我们选择了基于内核的虚拟机(KVM),因为它似乎是为基于arm的架构开发的最活跃的项目。本研究报告的指标包括TCP延迟、UDP延迟、TCP吞吐量和HTTP延迟。总的来说,ODROID-N2+的网络性能超过了Raspberry Pi 4 Model b。然而,后者比前者具有无可争辩的优势,因为它拥有更大、更活跃的社区,使得应用程序的开发和部署更快、更直接。因此,应该谨慎选择合适的sbc,考虑项目计划连接到sbc所需的软件和其他硬件。
Network Performance Evaluation Between Virtual/Native Nodes Running on ARM-based SBCs Using KVM as Hypervisor
Single-Board Computers (SBCs) are increasingly being used due to their small form factor, reduced energy consumption, versatility, affordability, and increasing computational power. Therefore, they are now used in projects where they were not initially contemplated, such as running a Virtual Machine Manager (VMM). In this work, two ARM-based SBCs were selected (Raspberry Pi 4 Model B and ODROID-N2+) and an empirical evaluation was carried out to evaluate the network performance between two nodes running in the same SBC, or in different SBCs directly connected through WiFi or Ethernet. Although several hypervisors are suitable for these SBCs, Kernel-based Virtual Machine (KVM) was chosen since it seems to be the most active project that is developed for the ARM-based architecture. The metrics reported in this study include the TCP latency, UDP latency, TCP throughput, and HTTP latency. In general, the network performance of the ODROID-N2+ exceeded the Raspberry Pi 4 Model B. However, the latter has an indisputable advantage over the former with a much larger and more active community, making the development and deployment of applications much faster and straightforward. Hence, selecting the suitable SBCs should be done cautiously, considering the required software and additional hardware that the project is planning to connect to the SBCs.