Abstract
In the modern era of computing, the cloud computing platform became popular with its on-demand resource scalability feature. Virtualization is the key technology to achieve resource scalability in the cloud environment. Virtual machine monitors (VMMs) like Xen and KVM are the enabling tools for virtualizing the resources in the cloud environment. The major role of VMMs is to map virtual CPUs of virtual machines to physical CPUs, popularly known as CPU scheduling. In this study, we analyzed the CPU scheduler of Xen VMM, called Credit CPU scheduler, with respect to CPU utilization. In order to maximize the physical CPU utilization in Xen VMM, the existing Credit CPU scheduling scheme distributes the physical CPU time among all the virtual CPUs available in virtual machines based on the current weight of the virtual machine. However, this scheduling approach wastes a considerable amount of CPU time in context switching due to random allocation of virtual CPUs to the real CPU cores. In addition to that, the Credit CPU scheduler in Xen is least concerned about load balancing at both virtual and physical CPUs level. Considering all these above issues, in this paper, a dynamic CPU scheduling algorithm is presented that will distribute both single and multithreaded load fairly among virtual and real CPUs, and also handle the issues related to context switching. The proposed CPU scheduling approach is implemented in Xen VMM using the virsh interface. The experimental results indicate that the proposed CPU scheduling approach distributes the available real CPU time evenly among virtual CPUs, which leads to balanced load in the Xen environment.
Highlights
Cloud computing is a modernistic computing paradigm based on the pay-per-use model
Procedure balance_vcpus() Let nCP U =number of real CPU cores Let max_vcpu_count=maximum number of virtual CPUs (vCPUs) pinned to any domain i Compute the current threshold curr_thr =100/ max_vcpu_count for i =1 to no_act_domains do if total_domain_load/nrV irtCpui > curr_thr and nrV irtCpui < nCP U pin one additional vCPU for domain i update the cpumapsi end if for j=1 to nrV irtCpui do if set bits in cpumapsi > 1 find the pCPU with minimum number of pinned vCPU pin the pCPU to the vCPU end if end for end for for i =1 to no_act_domains do find the underutilized vCPU unpin the vCPU from domain i end for Algorithm 2: Algorithm for load balancing at vCPU level
It is observed that the virtual machine instances built using the full virtualization technique are unable to adopt the dynamic changes in vCPUpCPU mapping for load balancing at vCPU as well as pCPU level
Summary
Cloud computing is a modernistic computing paradigm based on the pay-per-use model. Cloud consumers are required to pay as per their resource usability. The scalability feature of the Cloud supports contraction of capital costs by reducing the physical infrastructure. In order to achieve scalability, the cloud computing model makes use of virtualization technology. Virtualization [1] is the process of creating virtual machines over underlying hardware that simulates a real system. Virtual machine monitors (VMMs) like Xen [2], KVM [3], and Vmware are the key environments for virtualizing the physical systems into virtual systems. There is a need for proper virtual machine management that should maximize resource utilization and minimize the physical infrastructure. Virtual machine management is drawing the attention of researchers in the field of virtualization
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: TURKISH JOURNAL OF ELECTRICAL ENGINEERING & COMPUTER SCIENCES
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.
