TY - JOUR
T1 - Mitigation technique for performance degradation of virtual machine owing to GPU pass-through in fog computing
AU - Kang, Jihun
AU - Yu, Heonchang
N1 - Funding Information:
Manuscript received November 30, 2017. This work was supported by Institute for Information and communications Technology Promotion (IITP) grant funded by the Korea government(MSIT) (No. 2018-0-00480, Developing the edge cloud platform for the real time services based on the mobility of connected cars). The authors are with the Department of Computer Science and Engineering, Korea University, email: {k2j23h, yuhc}@korea.ac.kr. H. Yu is the corresponding author. Digital Object Identifier: 10.1109/JCN.2018.000038
Publisher Copyright:
© 2018 KICS.
PY - 2018/6
Y1 - 2018/6
N2 - As the size of data increases and computation becomes complicated in fog computing environments, the need for highperformance computation is increasing. One of the most popular ways to improve the performance of a virtual machine (VM) is to allocate a graphic processing unit (GPU) to the VM for supporting general purpose computing on graphic processing unit (GPGPU) operations. The direct pass-through, often used for GPUs in VMs, is popular in the cloud because VMs can use the full functionality of the GPU and experience virtually no performance degradation owing to virtualization. Direct pass-through is very useful for improving the performance of VMs. However, since the GPU usage time is not considered in the VM scheduler that operates based on the central processing unit (CPU) usage time of the VM, the VM performing the GPGPU operation degrades the performance of other VMs. In this paper, we analyze the effect of the VM performing the GPGPU operation (GPGPU-intensive VM) on other VMs through experiments. Then, we propose a method to mitigate the performance degradation of other VMs by dynamically allocating the resource usage time of the VM and preventing the priority preemption of the GPGPU-intensive VM.
AB - As the size of data increases and computation becomes complicated in fog computing environments, the need for highperformance computation is increasing. One of the most popular ways to improve the performance of a virtual machine (VM) is to allocate a graphic processing unit (GPU) to the VM for supporting general purpose computing on graphic processing unit (GPGPU) operations. The direct pass-through, often used for GPUs in VMs, is popular in the cloud because VMs can use the full functionality of the GPU and experience virtually no performance degradation owing to virtualization. Direct pass-through is very useful for improving the performance of VMs. However, since the GPU usage time is not considered in the VM scheduler that operates based on the central processing unit (CPU) usage time of the VM, the VM performing the GPGPU operation degrades the performance of other VMs. In this paper, we analyze the effect of the VM performing the GPGPU operation (GPGPU-intensive VM) on other VMs through experiments. Then, we propose a method to mitigate the performance degradation of other VMs by dynamically allocating the resource usage time of the VM and preventing the priority preemption of the GPGPU-intensive VM.
KW - Fog computing
KW - general purpose computing on graphic processing unit (GPGPU)
KW - performance isolation
KW - virtualization
UR - http://www.scopus.com/inward/record.url?scp=85052338825&partnerID=8YFLogxK
U2 - 10.1109/JCN.2018.000038
DO - 10.1109/JCN.2018.000038
M3 - Article
AN - SCOPUS:85052338825
SN - 1229-2370
VL - 20
SP - 257
EP - 265
JO - Journal of Communications and Networks
JF - Journal of Communications and Networks
IS - 3
M1 - 8437206
ER -