Abstract
Both cloud computing and desktop virtualization are based on massive virtual machines (VMs). Due to the conventional template-based creation and the limited bandwidth of disk I/O, booting up massive VMs is time consuming and lacks flexibility as well. However, current solutions put their focus on the design of the distributed storage for massive VMs' images. Considering all VMs' images should be read into the memory for booting up, the limited local disks' I/O bandwidth is actually the big challenge. To tackle this challenge, in this paper, we propose a scheme based on the minimum dataset (for booting up an operating system), which is shared by all the VMs created from different images. It can not only accelerate the boot-up progress of massive VMs via sparing I/O overhead, but also support the independent VM as well as the closed-source OS such as Windows. To demonstrate its efficiency, we also implement and deploy the system, namely Fast VM, on the Xen platform for Windows virtual machines with two types of VM formats, VHD and QCOW2. Experimental results show that our approach can significantly reduce the I/O overhead and even spare the booting up time up to 64.8%.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.