Asmaa Tellabi, Jaafar Fehmi, Sabri Abdelbast, C. Ruland
{"title":"On the Analysis of the Impact of Scheduling Plans in Safety Critical Requirements in Virtualization","authors":"Asmaa Tellabi, Jaafar Fehmi, Sabri Abdelbast, C. Ruland","doi":"10.1109/ICoCSec47621.2019.8971126","DOIUrl":null,"url":null,"abstract":"Over the years, virtualization has been deployed due to the benefits it offers, such as a better interoperability and improved performances. Two popular architectures of computers have been widely used in virtualization, x86 and ARM processors. The key component in virtualization is the hypervisor, which is a virtual machine manager that allows multiple guest operating systems (OS) to run on a single host system at the same time. Researchers and practitioners identified the challenge of enabling critical applications to share a common hardware platform without interfering one another as one of the main virtualization safety challenges. In this context, XtratuM was specified as a hypervisor designed to meet safety critical requirements in virtualization. In this paper, a comparison between XtratuM communication performances will be provided for ARM and x86 architectures. Different schedules will be given to partitions in order to examine the effect it has on each communication channel. As it will be demonstrated in this paper, Queuing messages (QC) are slower than Sampling Messages (SP), and this is caused by the internal structure and how messages are sent between partitions. As seen from tests, the version of XtratuM for x86 processors is faster than the one for ARM processors; this is also caused by the differences found in both architectures. In addition, a modified system’s architecture based on a previous architecture will be presented, which was modified in order to fit more the limits of XtratuM.","PeriodicalId":272402,"journal":{"name":"2019 International Conference on Cybersecurity (ICoCSec)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2019 International Conference on Cybersecurity (ICoCSec)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICoCSec47621.2019.8971126","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
Over the years, virtualization has been deployed due to the benefits it offers, such as a better interoperability and improved performances. Two popular architectures of computers have been widely used in virtualization, x86 and ARM processors. The key component in virtualization is the hypervisor, which is a virtual machine manager that allows multiple guest operating systems (OS) to run on a single host system at the same time. Researchers and practitioners identified the challenge of enabling critical applications to share a common hardware platform without interfering one another as one of the main virtualization safety challenges. In this context, XtratuM was specified as a hypervisor designed to meet safety critical requirements in virtualization. In this paper, a comparison between XtratuM communication performances will be provided for ARM and x86 architectures. Different schedules will be given to partitions in order to examine the effect it has on each communication channel. As it will be demonstrated in this paper, Queuing messages (QC) are slower than Sampling Messages (SP), and this is caused by the internal structure and how messages are sent between partitions. As seen from tests, the version of XtratuM for x86 processors is faster than the one for ARM processors; this is also caused by the differences found in both architectures. In addition, a modified system’s architecture based on a previous architecture will be presented, which was modified in order to fit more the limits of XtratuM.