Deep Reinforcement Learning-Based Flow Scheduling for Power-Efficient Data Center Networks

Zehua Guo

doi:10.1007/978-981-19-4874-9_4

Deep Reinforcement Learning-Based Flow Scheduling for Power-Efficient Data Center Networks

Zehua Guo^*

^*Corresponding author for this work

School of Automation

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

Abstract

In this chapter, we introduce SmartFCT, which employs DRL coupled with SDN to improve the power efficiency of DCNs and guarantee FCT. SmartFCT dynamically collects traffic distribution from switches to train its DRL model. The well-trained DRL agent of SmartFCT can quickly analyze the complicated traffic characteristics and adaptively generate an action for scheduling flows and deliberately configuring margins for different links. Following the generated action, flows are consolidated into a few of active links and switches to save power, and fine-grained margin configuration for active links avoids FCT violation of unexpected flow bursts.

Original language	English
Title of host publication	SpringerBriefs in Computer Science
Publisher	Springer
Pages	39-52
Number of pages	14
DOIs	https://doi.org/10.1007/978-981-19-4874-9_4
Publication status	Published - 2022

Publication series

Name	SpringerBriefs in Computer Science
ISSN (Print)	2191-5768
ISSN (Electronic)	2191-5776

Access to Document

10.1007/978-981-19-4874-9_4

Cite this

Guo, Z. (2022). Deep Reinforcement Learning-Based Flow Scheduling for Power-Efficient Data Center Networks. In SpringerBriefs in Computer Science (pp. 39-52). (SpringerBriefs in Computer Science). Springer. https://doi.org/10.1007/978-981-19-4874-9_4

@inbook{7ff5ba76b3be4e87a92723153adc883f,

title = "Deep Reinforcement Learning-Based Flow Scheduling for Power-Efficient Data Center Networks",

abstract = "In this chapter, we introduce SmartFCT, which employs DRL coupled with SDN to improve the power efficiency of DCNs and guarantee FCT. SmartFCT dynamically collects traffic distribution from switches to train its DRL model. The well-trained DRL agent of SmartFCT can quickly analyze the complicated traffic characteristics and adaptively generate an action for scheduling flows and deliberately configuring margins for different links. Following the generated action, flows are consolidated into a few of active links and switches to save power, and fine-grained margin configuration for active links avoids FCT violation of unexpected flow bursts.",

author = "Zehua Guo",

note = "Publisher Copyright: {\textcopyright} 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.",

year = "2022",

doi = "10.1007/978-981-19-4874-9_4",

language = "English",

series = "SpringerBriefs in Computer Science",

publisher = "Springer",

pages = "39--52",

booktitle = "SpringerBriefs in Computer Science",

address = "Germany",

}

TY - CHAP

T1 - Deep Reinforcement Learning-Based Flow Scheduling for Power-Efficient Data Center Networks

AU - Guo, Zehua

PY - 2022

Y1 - 2022

N2 - In this chapter, we introduce SmartFCT, which employs DRL coupled with SDN to improve the power efficiency of DCNs and guarantee FCT. SmartFCT dynamically collects traffic distribution from switches to train its DRL model. The well-trained DRL agent of SmartFCT can quickly analyze the complicated traffic characteristics and adaptively generate an action for scheduling flows and deliberately configuring margins for different links. Following the generated action, flows are consolidated into a few of active links and switches to save power, and fine-grained margin configuration for active links avoids FCT violation of unexpected flow bursts.

AB - In this chapter, we introduce SmartFCT, which employs DRL coupled with SDN to improve the power efficiency of DCNs and guarantee FCT. SmartFCT dynamically collects traffic distribution from switches to train its DRL model. The well-trained DRL agent of SmartFCT can quickly analyze the complicated traffic characteristics and adaptively generate an action for scheduling flows and deliberately configuring margins for different links. Following the generated action, flows are consolidated into a few of active links and switches to save power, and fine-grained margin configuration for active links avoids FCT violation of unexpected flow bursts.

UR - http://www.scopus.com/inward/record.url?scp=85139854912&partnerID=8YFLogxK

U2 - 10.1007/978-981-19-4874-9_4

DO - 10.1007/978-981-19-4874-9_4

M3 - Chapter

AN - SCOPUS:85139854912

T3 - SpringerBriefs in Computer Science

SP - 39

EP - 52

BT - SpringerBriefs in Computer Science

PB - Springer

ER -

Deep Reinforcement Learning-Based Flow Scheduling for Power-Efficient Data Center Networks

Abstract

Publication series

Access to Document

Other files and links

Fingerprint

Cite this