In this paper the authors evaluate the CPU Consumption, Memory Utilization and Transfer Time between 5 Hypervisors XEN-PV, XEN-HVM, OpenVZ, KVM-FV and KVM-PV by using FTP and HTTP approaches. All the results are compared with the Real Environment. From the experimental results, the authors have concluded that OpenVZ and XEN-PV have better performance than other hypervisors. The worse performance is for KVM Hypervisor. In our paper we have used some scripts in order to evaluate the performance of these Hypervisors. Keyword: XEN-PV, XEN-HVM, OpenVZ, KVM-FV, KVM-PV, FTP approach, HTTP approach. 2011 Third International Conference on Intelligent Networking and Collaborative Systems 978-0-7695-4579-0/11 $26.00
Abstract-The aim of this paper is to compare the performance between three hypervisors: XEN-PV, XEN-HVM and Open-VZ. We have simulated the migration of a virtual machine by using a warning failure approach. Based on some experiments we have compared CPU Consumption, Memory Utilization, Total Migration Time and Downtime. We have also tested the hypervisor's performance by changing the packet's size from 1500 byte to 64 byte. From these tests we have concluded that Open-VZ has a bigger CPU Consumption than XEN-PV, but the Total Migration time is smaller than in XEN-PV. XEN-HVM has a worse performance than XEN-PV, especially regarding to Downtime parameter.
This paper presents a new solution, which adds a new database server on an existing Oracle MultimasterData replication system with Online Instantiation method. During this time the system is down, because we cannot execute DML statements on replication objects but we can only make queries. The time for adding the new database server depends on the number of objects, on the replication group and on the network conditions. We propose to add a new layer between replication objects and the database sessions, which contain DML statements. The layer eliminates the system down time exploiting our developed packages. The packages will be active only during the addition of a new site process and will modify all DML statements and queries based on replication objects.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.