SMQoS: Improving Utilization and Energy Efficiency with QoS Awareness on GPUs

Qingxiao Sun; Yi Liu; Hailong Yang; Zhongzhi Luan; Depei Qian

doi:10.1109/CLUSTER.2019.8891047

SMQoS: Improving Utilization and Energy Efficiency with QoS Awareness on GPUs

Qingxiao Sun
, Yi Liu
, Hailong Yang^*
, Zhongzhi Luan
, Depei Qian

^*Corresponding author for this work

School of Computer Science and Engineering

Beihang University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

9 Scopus citations

Abstract

Meeting the Quality of Service (QoS) requirement under task consolidation on the GPU is extremely challenging. Previous work mostly relies on static task or resource scheduling and cannot handle the QoS violation during runtime. In addition, the existing work fails to exploit the computing characteristics of batch tasks, and thus wastes the opportunities to reduce power consumption while improving GPU utilization. To address the above problems, we propose a new runtime mechanism SMQoS that can dynamically adjust the resource allocation during runtime to satisfy the QoS of latency-sensitive tasks and determine the optimal resource allocation for batch tasks to improve GPU utilization and power efficiency. The experimental results show that with SMQoS, 2.27% and 7.58% more task co-runnings reach the 95% QoS target than Spart and Rollover respectively. In addition, SMQoS achieves 23.9% and 32.3% higher throughput, and reduces the power consumption by 25.7% and 10.1%, compared to Spart and Rollover respectively.

Original language	English
Title of host publication	Proceedings - 2019 IEEE International Conference on Cluster Computing, CLUSTER 2019
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781728147345
DOIs	https://doi.org/10.1109/CLUSTER.2019.8891047
State	Published - Sep 2019
Event	2019 IEEE International Conference on Cluster Computing, CLUSTER 2019 - Albuquerque, United States Duration: 23 Sep 2019 → 26 Sep 2019

Publication series

Name	Proceedings - IEEE International Conference on Cluster Computing, ICCC
Volume	2019-September
ISSN (Print)	1552-5244

Conference

Conference	2019 IEEE International Conference on Cluster Computing, CLUSTER 2019
Country/Territory	United States
City	Albuquerque
Period	23/09/19 → 26/09/19

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 7 Affordable and Clean Energy

Keywords

Dynamic resource management
Graphics processing units
Power efficiency
Quality of Service
Throughput

Access to Document

10.1109/CLUSTER.2019.8891047

Cite this

Sun, Q., Liu, Y., Yang, H., Luan, Z., & Qian, D. (2019). SMQoS: Improving Utilization and Energy Efficiency with QoS Awareness on GPUs. In Proceedings - 2019 IEEE International Conference on Cluster Computing, CLUSTER 2019 Article 8891047 (Proceedings - IEEE International Conference on Cluster Computing, ICCC; Vol. 2019-September). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/CLUSTER.2019.8891047

@inproceedings{6531afaa391a4332b1736f8b1006a929,

title = "SMQoS: Improving Utilization and Energy Efficiency with QoS Awareness on GPUs",

abstract = "Meeting the Quality of Service (QoS) requirement under task consolidation on the GPU is extremely challenging. Previous work mostly relies on static task or resource scheduling and cannot handle the QoS violation during runtime. In addition, the existing work fails to exploit the computing characteristics of batch tasks, and thus wastes the opportunities to reduce power consumption while improving GPU utilization. To address the above problems, we propose a new runtime mechanism SMQoS that can dynamically adjust the resource allocation during runtime to satisfy the QoS of latency-sensitive tasks and determine the optimal resource allocation for batch tasks to improve GPU utilization and power efficiency. The experimental results show that with SMQoS, 2.27\% and 7.58\% more task co-runnings reach the 95\% QoS target than Spart and Rollover respectively. In addition, SMQoS achieves 23.9\% and 32.3\% higher throughput, and reduces the power consumption by 25.7\% and 10.1\%, compared to Spart and Rollover respectively.",

keywords = "Dynamic resource management, Graphics processing units, Power efficiency, Quality of Service, Throughput",

author = "Qingxiao Sun and Yi Liu and Hailong Yang and Zhongzhi Luan and Depei Qian",

note = "Publisher Copyright: {\textcopyright} 2019 IEEE.; 2019 IEEE International Conference on Cluster Computing, CLUSTER 2019 ; Conference date: 23-09-2019 Through 26-09-2019",

year = "2019",

month = sep,

doi = "10.1109/CLUSTER.2019.8891047",

language = "英语",

series = "Proceedings - IEEE International Conference on Cluster Computing, ICCC",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "Proceedings - 2019 IEEE International Conference on Cluster Computing, CLUSTER 2019",

address = "美国",

}

Sun, Q, Liu, Y , Yang, H , Luan, Z & Qian, D 2019, SMQoS: Improving Utilization and Energy Efficiency with QoS Awareness on GPUs. in Proceedings - 2019 IEEE International Conference on Cluster Computing, CLUSTER 2019., 8891047, Proceedings - IEEE International Conference on Cluster Computing, ICCC, vol. 2019-September, Institute of Electrical and Electronics Engineers Inc., 2019 IEEE International Conference on Cluster Computing, CLUSTER 2019, Albuquerque, United States, 23/09/19. https://doi.org/10.1109/CLUSTER.2019.8891047

SMQoS: Improving Utilization and Energy Efficiency with QoS Awareness on GPUs. / Sun, Qingxiao; Liu, Yi ; Yang, Hailong et al.
Proceedings - 2019 IEEE International Conference on Cluster Computing, CLUSTER 2019. Institute of Electrical and Electronics Engineers Inc., 2019. 8891047 (Proceedings - IEEE International Conference on Cluster Computing, ICCC; Vol. 2019-September).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - SMQoS

T2 - 2019 IEEE International Conference on Cluster Computing, CLUSTER 2019

AU - Sun, Qingxiao

AU - Liu, Yi

AU - Yang, Hailong

AU - Luan, Zhongzhi

AU - Qian, Depei

PY - 2019/9

Y1 - 2019/9

N2 - Meeting the Quality of Service (QoS) requirement under task consolidation on the GPU is extremely challenging. Previous work mostly relies on static task or resource scheduling and cannot handle the QoS violation during runtime. In addition, the existing work fails to exploit the computing characteristics of batch tasks, and thus wastes the opportunities to reduce power consumption while improving GPU utilization. To address the above problems, we propose a new runtime mechanism SMQoS that can dynamically adjust the resource allocation during runtime to satisfy the QoS of latency-sensitive tasks and determine the optimal resource allocation for batch tasks to improve GPU utilization and power efficiency. The experimental results show that with SMQoS, 2.27% and 7.58% more task co-runnings reach the 95% QoS target than Spart and Rollover respectively. In addition, SMQoS achieves 23.9% and 32.3% higher throughput, and reduces the power consumption by 25.7% and 10.1%, compared to Spart and Rollover respectively.

AB - Meeting the Quality of Service (QoS) requirement under task consolidation on the GPU is extremely challenging. Previous work mostly relies on static task or resource scheduling and cannot handle the QoS violation during runtime. In addition, the existing work fails to exploit the computing characteristics of batch tasks, and thus wastes the opportunities to reduce power consumption while improving GPU utilization. To address the above problems, we propose a new runtime mechanism SMQoS that can dynamically adjust the resource allocation during runtime to satisfy the QoS of latency-sensitive tasks and determine the optimal resource allocation for batch tasks to improve GPU utilization and power efficiency. The experimental results show that with SMQoS, 2.27% and 7.58% more task co-runnings reach the 95% QoS target than Spart and Rollover respectively. In addition, SMQoS achieves 23.9% and 32.3% higher throughput, and reduces the power consumption by 25.7% and 10.1%, compared to Spart and Rollover respectively.

KW - Dynamic resource management

KW - Graphics processing units

KW - Power efficiency

KW - Quality of Service

KW - Throughput

UR - https://www.scopus.com/pages/publications/85075270219

U2 - 10.1109/CLUSTER.2019.8891047

DO - 10.1109/CLUSTER.2019.8891047

M3 - 会议稿件

AN - SCOPUS:85075270219

T3 - Proceedings - IEEE International Conference on Cluster Computing, ICCC

BT - Proceedings - 2019 IEEE International Conference on Cluster Computing, CLUSTER 2019

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 23 September 2019 through 26 September 2019

ER -

Sun Q, Liu Y , Yang H , Luan Z , Qian D. SMQoS: Improving Utilization and Energy Efficiency with QoS Awareness on GPUs. In Proceedings - 2019 IEEE International Conference on Cluster Computing, CLUSTER 2019. Institute of Electrical and Electronics Engineers Inc. 2019. 8891047. (Proceedings - IEEE International Conference on Cluster Computing, ICCC). doi: 10.1109/CLUSTER.2019.8891047

SMQoS: Improving Utilization and Energy Efficiency with QoS Awareness on GPUs

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this