Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

DDPC : Automated Data-Driven Power-Performance Controller Design on-the-fly for Latency-sensitive Web Services

Savasci, Mehmet ; Ali-Eldin, Ahmed ; Eker, Johan LU orcid ; Robertsson, Anders LU and Shenoy, Prashant (2023) 2023 World Wide Web Conference, WWW 2023 In ACM Web Conference 2023 - Proceedings of the World Wide Web Conference, WWW 2023 p.3067-3076
Abstract

Traditional power reduction techniques such as DVFS or RAPL are challenging to use with web services because they significantly affect the services' latency and throughput. Previous work suggested the use of controllers based on control theory or machine learning to reduce performance degradation under constrained power. However, generating these controllers is challenging as every web service applications running in a data center requires a power-performance model and a fine-tuned controller. In this paper, we present DDPC, a system for autonomic data-driven controller generation for power-latency management. DDPC automates the process of designing and deploying controllers for dynamic power allocation to manage the power-performance... (More)

Traditional power reduction techniques such as DVFS or RAPL are challenging to use with web services because they significantly affect the services' latency and throughput. Previous work suggested the use of controllers based on control theory or machine learning to reduce performance degradation under constrained power. However, generating these controllers is challenging as every web service applications running in a data center requires a power-performance model and a fine-tuned controller. In this paper, we present DDPC, a system for autonomic data-driven controller generation for power-latency management. DDPC automates the process of designing and deploying controllers for dynamic power allocation to manage the power-performance trade-offs for latency-sensitive web applications such as a social network. For each application, DDPC uses system identification techniques to learn an adaptive power-performance model that captures the application's power-latency trade-offs which is then used to generate and deploy a Proportional-Integral (PI) power controller with gain-scheduling to dynamically manage the power allocation to the server running application using RAPL. We evaluate DDPC with two realistic latency-sensitive web applications under varying load scenarios. Our results show that DDPC is capable of autonomically generating and deploying controllers within a few minutes reducing the active power allocation of a web-server by more than 50% compared to state-of-the-art techniques while maintaining the latency well below the target of the application.

(Less)
Please use this url to cite or link to this publication:
author
; ; ; and
organization
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
keywords
datacenter, power-management, Web service performance
host publication
ACM Web Conference 2023 - Proceedings of the World Wide Web Conference, WWW 2023
series title
ACM Web Conference 2023 - Proceedings of the World Wide Web Conference, WWW 2023
pages
10 pages
publisher
Association for Computing Machinery (ACM)
conference name
2023 World Wide Web Conference, WWW 2023
conference location
Austin, United States
conference dates
2023-04-30 - 2023-05-04
external identifiers
  • scopus:85159298066
ISBN
9781450394161
DOI
10.1145/3543507.3583437
language
English
LU publication?
yes
id
295d9fbb-f5eb-47f4-a1ed-f6238e414c66
date added to LUP
2023-08-11 12:58:28
date last changed
2023-11-22 20:56:59
@inproceedings{295d9fbb-f5eb-47f4-a1ed-f6238e414c66,
  abstract     = {{<p>Traditional power reduction techniques such as DVFS or RAPL are challenging to use with web services because they significantly affect the services' latency and throughput. Previous work suggested the use of controllers based on control theory or machine learning to reduce performance degradation under constrained power. However, generating these controllers is challenging as every web service applications running in a data center requires a power-performance model and a fine-tuned controller. In this paper, we present DDPC, a system for autonomic data-driven controller generation for power-latency management. DDPC automates the process of designing and deploying controllers for dynamic power allocation to manage the power-performance trade-offs for latency-sensitive web applications such as a social network. For each application, DDPC uses system identification techniques to learn an adaptive power-performance model that captures the application's power-latency trade-offs which is then used to generate and deploy a Proportional-Integral (PI) power controller with gain-scheduling to dynamically manage the power allocation to the server running application using RAPL. We evaluate DDPC with two realistic latency-sensitive web applications under varying load scenarios. Our results show that DDPC is capable of autonomically generating and deploying controllers within a few minutes reducing the active power allocation of a web-server by more than 50% compared to state-of-the-art techniques while maintaining the latency well below the target of the application.</p>}},
  author       = {{Savasci, Mehmet and Ali-Eldin, Ahmed and Eker, Johan and Robertsson, Anders and Shenoy, Prashant}},
  booktitle    = {{ACM Web Conference 2023 - Proceedings of the World Wide Web Conference, WWW 2023}},
  isbn         = {{9781450394161}},
  keywords     = {{datacenter; power-management; Web service performance}},
  language     = {{eng}},
  month        = {{04}},
  pages        = {{3067--3076}},
  publisher    = {{Association for Computing Machinery (ACM)}},
  series       = {{ACM Web Conference 2023 - Proceedings of the World Wide Web Conference, WWW 2023}},
  title        = {{DDPC : Automated Data-Driven Power-Performance Controller Design on-the-fly for Latency-sensitive Web Services}},
  url          = {{http://dx.doi.org/10.1145/3543507.3583437}},
  doi          = {{10.1145/3543507.3583437}},
  year         = {{2023}},
}