Improving the management efficiency of GPU workloads in data centers through GPU virtualization
View/ Open
Impact
Scholar |
Other documents of the author: Iserte, Sergio; Prades, Javier; Reaño, Carlos; Silla, Federico
Metadata
Show full item recordcomunitat-uji-handle:10234/9
comunitat-uji-handle2:10234/7036
comunitat-uji-handle3:10234/8620
comunitat-uji-handle4:
INVESTIGACIONMetadata
Title
Improving the management efficiency of GPU workloads in data centers through GPU virtualizationDate
2019Publisher
WileyISSN
1532-0626; 1532-0634Bibliographic citation
Iserte, S, Prades, J, Reaño, C, Silla, F. Improving the management efficiency of GPU workloads in data centers through GPU virtualization. Concurrency Computat Pract Exper. 2019;e5275. https://doi.org/10.1002/cpe.5275Type
info:eu-repo/semantics/articlePublisher version
https://onlinelibrary.wiley.com/doi/full/10.1002/cpe.5275Version
info:eu-repo/semantics/submittedVersionSubject
Abstract
Graphics processing units (GPUs) are currently used in data centers to reduce the execution time of compute‐intensive applications. However, the use of GPUs presents several side effects, such as increased acquisition ... [+]
Graphics processing units (GPUs) are currently used in data centers to reduce the execution time of compute‐intensive applications. However, the use of GPUs presents several side effects, such as increased acquisition costs and larger space requirements. Furthermore, GPUs require a nonnegligible amount of energy even while idle. Additionally, GPU utilization is usually low for most applications. In a similar way to the use of virtual machines, using virtual GPUs may address the concerns associated with the use of these devices. In this regard, the remote GPU virtualization mechanism could be leveraged to share the GPUs present in the computing facility among the nodes of the cluster. This would increase overall GPU utilization, thus reducing the negative impact of the increased costs mentioned before. Reducing the amount of GPUs installed in the cluster could also be possible. However, in the same way as job schedulers map GPU resources to applications, virtual GPUs should also be scheduled before job execution. Nevertheless, current job schedulers are not able to deal with virtual GPUs. In this paper, we analyze the performance attained by a cluster using the remote Compute Unified Device Architecture middleware and a modified version of the Slurm scheduler, which is now able to assign remote GPUs to jobs. Results show that cluster throughput, measured as jobs completed per time unit, is doubled at the same time that the total energy consumption is reduced up to 40%. GPU utilization is also increased. [-]
Description
This is the pre-peer reviewed version of the following article: Improving the management efficiency of GPU workloads in data centers through GPU virtualization, which has been published in final form at https://doi. ... [+]
This is the pre-peer reviewed version of the following article: Improving the management efficiency of GPU workloads in data centers through GPU virtualization, which has been published in final form at https://doi.org/10.1002/cpe.5275. This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for Use of Self-Archived Versions. [-]
Is part of
Practice and Experience, 2019Investigation project
Generalitat Valenciana. Grant Number: PROMETEO/2017/077; MINECO and FEDER. Grant Numbers: TIN2014-53495-R, TIN2015-65316-P, TIN2017-82972-RRights
Copyright © John Wiley & Sons, Inc.
http://rightsstatements.org/vocab/InC/1.0/
info:eu-repo/semantics/openAccess
http://rightsstatements.org/vocab/InC/1.0/
info:eu-repo/semantics/openAccess
This item appears in the folowing collection(s)
- ICC_Articles [419]