Dynamic flow scheduling for Power-efficient Data Center Networks

Zehua Guo, Shufeng Hui, Yang Xu, H. Jonathan Chao

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Power-efficient Data Center Networks (DCNs) have been proposed to save power of DCNs using OpenFlow. In these DCNs, the OpenFlow controller adaptively turns on and off links and OpenFlow switches to form a minimum-power subnet that satisfies traffic demand. As the subnet changes, flows are scheduled dynamically to routes composed of active switches and links. However, existing flow scheduling schemes could cause undesired results: (1) power inefficiency: due to unbalanced traffic allocation on active routes, extra switches and links may be activated to cater to bursty traffic surges on congested routes, and (2) Quality of Service (QoS) fluctuation: because of the limited flow entry processing ability, switches cannot timely install/delete/update flow entries to properly schedule flows. In this paper, we propose AggreFlow, a dynamic flow scheduling scheme that achieves power efficiency in DCNs and improved QoS using two techniques: Flow-set Routing and Lazy Rerouting. Flow-set Routing achieves load balancing and reduces the number of entry installment on switches by routing flows in a coarse-grained flow-set fashion. Lazy Rerouting maintains load balancing and spreads rerouting operations over a relatively long period of time, reducing the burstiness of entry installment/deletion/update on switches. We built a NS3 based fat-tree network simulation platform to evaluate AggreFlow's performance. The simulation results show AggreFlow reduces power consumption by about 18%, achieves load balancing and improved QoS (i.e., low packet loss rate and reducing the number of processing entries for flow scheduling by 98%), compared with baseline schemes.

Original languageEnglish (US)
Title of host publication2016 IEEE/ACM 24th International Symposium on Quality of Service, IWQoS 2016
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781509026340
DOIs
StatePublished - Oct 13 2016
Event24th IEEE/ACM International Symposium on Quality of Service, IWQoS 2016 - Beijing, China
Duration: Jun 20 2016Jun 21 2016

Other

Other24th IEEE/ACM International Symposium on Quality of Service, IWQoS 2016
CountryChina
CityBeijing
Period6/20/166/21/16

Fingerprint

Scheduling
Switches
Resource allocation
Quality of service
Telecommunication traffic
Packet loss
Processing
Oils and fats
Telecommunication links
Data center
Electric power utilization
Controllers
Routing
Load balancing

Keywords

  • Flow scheduling
  • OpenFlow
  • Power saving
  • Power-efficient data center networks

ASJC Scopus subject areas

  • Management of Technology and Innovation
  • Computer Networks and Communications
  • Media Technology
  • Safety, Risk, Reliability and Quality

Cite this

Guo, Z., Hui, S., Xu, Y., & Chao, H. J. (2016). Dynamic flow scheduling for Power-efficient Data Center Networks. In 2016 IEEE/ACM 24th International Symposium on Quality of Service, IWQoS 2016 [7590399] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IWQoS.2016.7590399

Dynamic flow scheduling for Power-efficient Data Center Networks. / Guo, Zehua; Hui, Shufeng; Xu, Yang; Chao, H. Jonathan.

2016 IEEE/ACM 24th International Symposium on Quality of Service, IWQoS 2016. Institute of Electrical and Electronics Engineers Inc., 2016. 7590399.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Guo, Z, Hui, S, Xu, Y & Chao, HJ 2016, Dynamic flow scheduling for Power-efficient Data Center Networks. in 2016 IEEE/ACM 24th International Symposium on Quality of Service, IWQoS 2016., 7590399, Institute of Electrical and Electronics Engineers Inc., 24th IEEE/ACM International Symposium on Quality of Service, IWQoS 2016, Beijing, China, 6/20/16. https://doi.org/10.1109/IWQoS.2016.7590399
Guo Z, Hui S, Xu Y, Chao HJ. Dynamic flow scheduling for Power-efficient Data Center Networks. In 2016 IEEE/ACM 24th International Symposium on Quality of Service, IWQoS 2016. Institute of Electrical and Electronics Engineers Inc. 2016. 7590399 https://doi.org/10.1109/IWQoS.2016.7590399
Guo, Zehua ; Hui, Shufeng ; Xu, Yang ; Chao, H. Jonathan. / Dynamic flow scheduling for Power-efficient Data Center Networks. 2016 IEEE/ACM 24th International Symposium on Quality of Service, IWQoS 2016. Institute of Electrical and Electronics Engineers Inc., 2016.
@inproceedings{9da3ff1e9f7b47238dea10fe5c04cfd0,
title = "Dynamic flow scheduling for Power-efficient Data Center Networks",
abstract = "Power-efficient Data Center Networks (DCNs) have been proposed to save power of DCNs using OpenFlow. In these DCNs, the OpenFlow controller adaptively turns on and off links and OpenFlow switches to form a minimum-power subnet that satisfies traffic demand. As the subnet changes, flows are scheduled dynamically to routes composed of active switches and links. However, existing flow scheduling schemes could cause undesired results: (1) power inefficiency: due to unbalanced traffic allocation on active routes, extra switches and links may be activated to cater to bursty traffic surges on congested routes, and (2) Quality of Service (QoS) fluctuation: because of the limited flow entry processing ability, switches cannot timely install/delete/update flow entries to properly schedule flows. In this paper, we propose AggreFlow, a dynamic flow scheduling scheme that achieves power efficiency in DCNs and improved QoS using two techniques: Flow-set Routing and Lazy Rerouting. Flow-set Routing achieves load balancing and reduces the number of entry installment on switches by routing flows in a coarse-grained flow-set fashion. Lazy Rerouting maintains load balancing and spreads rerouting operations over a relatively long period of time, reducing the burstiness of entry installment/deletion/update on switches. We built a NS3 based fat-tree network simulation platform to evaluate AggreFlow's performance. The simulation results show AggreFlow reduces power consumption by about 18{\%}, achieves load balancing and improved QoS (i.e., low packet loss rate and reducing the number of processing entries for flow scheduling by 98{\%}), compared with baseline schemes.",
keywords = "Flow scheduling, OpenFlow, Power saving, Power-efficient data center networks",
author = "Zehua Guo and Shufeng Hui and Yang Xu and Chao, {H. Jonathan}",
year = "2016",
month = "10",
day = "13",
doi = "10.1109/IWQoS.2016.7590399",
language = "English (US)",
booktitle = "2016 IEEE/ACM 24th International Symposium on Quality of Service, IWQoS 2016",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
address = "United States",

}

TY - GEN

T1 - Dynamic flow scheduling for Power-efficient Data Center Networks

AU - Guo, Zehua

AU - Hui, Shufeng

AU - Xu, Yang

AU - Chao, H. Jonathan

PY - 2016/10/13

Y1 - 2016/10/13

N2 - Power-efficient Data Center Networks (DCNs) have been proposed to save power of DCNs using OpenFlow. In these DCNs, the OpenFlow controller adaptively turns on and off links and OpenFlow switches to form a minimum-power subnet that satisfies traffic demand. As the subnet changes, flows are scheduled dynamically to routes composed of active switches and links. However, existing flow scheduling schemes could cause undesired results: (1) power inefficiency: due to unbalanced traffic allocation on active routes, extra switches and links may be activated to cater to bursty traffic surges on congested routes, and (2) Quality of Service (QoS) fluctuation: because of the limited flow entry processing ability, switches cannot timely install/delete/update flow entries to properly schedule flows. In this paper, we propose AggreFlow, a dynamic flow scheduling scheme that achieves power efficiency in DCNs and improved QoS using two techniques: Flow-set Routing and Lazy Rerouting. Flow-set Routing achieves load balancing and reduces the number of entry installment on switches by routing flows in a coarse-grained flow-set fashion. Lazy Rerouting maintains load balancing and spreads rerouting operations over a relatively long period of time, reducing the burstiness of entry installment/deletion/update on switches. We built a NS3 based fat-tree network simulation platform to evaluate AggreFlow's performance. The simulation results show AggreFlow reduces power consumption by about 18%, achieves load balancing and improved QoS (i.e., low packet loss rate and reducing the number of processing entries for flow scheduling by 98%), compared with baseline schemes.

AB - Power-efficient Data Center Networks (DCNs) have been proposed to save power of DCNs using OpenFlow. In these DCNs, the OpenFlow controller adaptively turns on and off links and OpenFlow switches to form a minimum-power subnet that satisfies traffic demand. As the subnet changes, flows are scheduled dynamically to routes composed of active switches and links. However, existing flow scheduling schemes could cause undesired results: (1) power inefficiency: due to unbalanced traffic allocation on active routes, extra switches and links may be activated to cater to bursty traffic surges on congested routes, and (2) Quality of Service (QoS) fluctuation: because of the limited flow entry processing ability, switches cannot timely install/delete/update flow entries to properly schedule flows. In this paper, we propose AggreFlow, a dynamic flow scheduling scheme that achieves power efficiency in DCNs and improved QoS using two techniques: Flow-set Routing and Lazy Rerouting. Flow-set Routing achieves load balancing and reduces the number of entry installment on switches by routing flows in a coarse-grained flow-set fashion. Lazy Rerouting maintains load balancing and spreads rerouting operations over a relatively long period of time, reducing the burstiness of entry installment/deletion/update on switches. We built a NS3 based fat-tree network simulation platform to evaluate AggreFlow's performance. The simulation results show AggreFlow reduces power consumption by about 18%, achieves load balancing and improved QoS (i.e., low packet loss rate and reducing the number of processing entries for flow scheduling by 98%), compared with baseline schemes.

KW - Flow scheduling

KW - OpenFlow

KW - Power saving

KW - Power-efficient data center networks

UR - http://www.scopus.com/inward/record.url?scp=85009816958&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85009816958&partnerID=8YFLogxK

U2 - 10.1109/IWQoS.2016.7590399

DO - 10.1109/IWQoS.2016.7590399

M3 - Conference contribution

BT - 2016 IEEE/ACM 24th International Symposium on Quality of Service, IWQoS 2016

PB - Institute of Electrical and Electronics Engineers Inc.

ER -