An SDN Focused Approach for Energy Aware Traffic Engineering in Data Centers

There is a lot of effort to limit the impact of CO2 emissions from the information communication technologies (ICT) industry by reducing the energy consumption on all aspects of networking technologies. In a service provider network, data centers (DCs) are the major power consumer and considerable g...

Full description

Saved in:
Bibliographic Details
Published in:Sensors (Basel, Switzerland) Switzerland), 2019-09, Vol.19 (18), p.3980
Main Authors: Charalampou, Paris, Sykas, Efstathios D.
Format: Article
Language:English
Subjects:
Architectural engineering
Climate change
Computer centers
Computer simulation
data center
Data centers
Energy
energy awareness
Energy consumption
Engineering
Integer programming
Internet of Things
Internet service providers
Manufacturers
Network topologies
optimization
Optimization algorithms
Power management
Software
software defined networking (SDN)
Software-defined networking
Traffic congestion
Traffic control
Traffic engineering
Wireless networks
Workloads
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:There is a lot of effort to limit the impact of CO2 emissions from the information communication technologies (ICT) industry by reducing the energy consumption on all aspects of networking technologies. In a service provider network, data centers (DCs) are the major power consumer and considerable gains are expected by regulating the operation network devices. In this context, we developed a mixed integer programming (MIP) algorithm to optimize the power consumption of network devices via energy aware traffic engineering. We verified our approach by simulating DC network topologies and demonstrated that clear benefits can be achieved for various network sizes and traffic volumes. Our algorithm can be easily implemented as an application in the software-defined networking (SDN) paradigm, making quite feasible its deployment in a production environment.
ISSN:1424-8220
1424-8220
DOI:10.3390/s19183980