ERTDS: A dynamic CPU scheduler for Xen virtualization systems
Jun Wu, Jianfu Li
Abstract
Jun Wu, Jianfu Li
Abstract
Real-time deferrable server (RTDS) scheduler is presented since Xen 4.5. Under RTDS, a guaranteed physical CPU capacity is provided to every virtual CPU so that the performance can be better predicted. However, the guaranteed capacity is defined off-line, it might not fit the requirement of a virtual CPU at the run-time. In this paper, an RTDS-based CPU scheduler is proposed, called enhanced real-time deferrable server (ERTDS), to provide an additional capacity to virtual CPUs when their run-time requirements are higher than expected. The proposed ERTDS has been implemented in Xen 4.7 and a series of experiments has been conducted for which we have some encouraging results.
OpenAlex reports 3 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Real-time deferrable server (RTDS) scheduler is presented since Xen 4.5. Under RTDS, a guaranteed physical CPU capacity is provided to every virtual CPU so that the performance can be better predicted. However, the guaranteed capacity is defined off-line, it might not fit the requirement of a virtual CPU at the run-time. In this paper, an RTDS-based CPU scheduler is proposed, called enhanced real-time deferrable server (ERTDS), to provide an additional capacity to virtual CPUs when their run-time requirements are higher than expected. The proposed ERTDS has been implemented in Xen 4.7 and a series of experiments has been conducted for which we have some encouraging results.
Key concepts: Virtualization, Computer science, Operating system, CPU shielding, Central processing unit, Virtual machine, Embedded system, Cloud computing