Search

Virtual machines are used in various fields to increase the resource utilization. In a virtual machine monitor, I/O requests are the most frequent tasks for virtual machines. However, a virtual machine monitor cannot guarantee I/O responsiveness when multiple tasks are being performed on virtual machines. We propose a scheduling scheme that provides fast I/O responsiveness for virtual machines. The proposed scheme provides high resource utilization by dynamically allocating virtual resources to virtual machines according to the characteristics of their tasks. In experiments with virtual machines containing varying task characteristics, it is shown that the proposed scheme provides improved I/O responsiveness when compared to existing virtual machine schedulers.

Credit scheduler, an original Xen virtual machine scheduler, uses a boost mechanism to improve responsiveness. However, the original boost mechanism is applied only to inactive virtual machines. Because of this, if virtual machine scheduling delay time increases, the I/O performance of an active virtual machine dramatically decreases. However, when the boost mechanism of a previous credit scheduler is applied to an active virtual machine, the allocation equality can become unbalanced. This research study proposes a boost mechanism that can be applied to an active virtual machine. Evaluation of the results showed that, when compared to the previous credit scheduler, the proposed method both did not upset the fairness of the CPU allocation between the virtual machines and improved the I/O performance.

Through recent developments in hardware, it has become possible to interpret user intent by integrating biological information recorded through various biometric devices. However, because multiple devices that acquire biological information have different OSs, data transmission protocols, and arrival times, it is not easy to collect and analyze biological information in real-time. Our proposed method solves the issue of different OS and data transmission protocols. To realize the same, the proposed synchronization framework implements a biometric information priority and algorithm. The implemented framework is able to analyze and process data in real-time in order to recognize user intention information.

This paper exposes the original Xen credit scheduler. In some case, the original credit scheduler results in low responsiveness. The credit scheduler is a proportional fair share central processing unit (CPU) scheduler and provides CPU fairness. However, in contrast with fine CPU fairness, the responsiveness is insufficient in the case of CPU intensive workloads. We modify the Xen credit scheduler to obtain an improved response time because VM scheduling significantly affects the responsiveness of virtual machines (VMs) in virtualization environments. The supposed scheduling boosts a VM with a conditional boost mechanism to maintain CPU fairness. The experimental results show that the supposed scheduling technique offers better responsiveness for a VM than the original credit scheduler.