Please use this identifier to cite or link to this item:
Title: iPRDR: Intelligent Power Reduction Decision Routing Protocol for Big Traffic Flood in Hybrid-SDN Architecture
Authors: Al-Mhdawi, A
Al-Raweshidy, H
Keywords: Power reduction decision routing protocol;Performance index;Layer metric;Uplink utilization;Power consumption model;OF switches
Issue Date: 2018
Citation: IEEE Access
Abstract: Analysing data centres energy consumption is the main step towards building a reliable infrastructure. Evidently, data centers consume a large number of billions of gigabytes information to the point that putting tremendous pressure on energy suppliers. Every internet activity involves a huge amount of data that need to be stored in a cloud data center somewhere, not forgetting the internet of things (IoT) applications and other social media services that produce an extraordinarily large scale of big data that require high processing and analysis. Moreover, current Data centres consume about 3% of global electricity supply which is about 416.2 TWh of power that world data centers consumed in 2016. In this paper, we have developed an Intelligent Power Reduction Decision Routing Protocol (iPRDR) in a medium scale Hybrid-Software Defined Network (H-SDN) data centre environment. The proposed iPRDR protocol approach is to dynamically segregate big traffic and route it to a high index processing devices with a power-optimal selected path. The protocol approach is to decrease the overall power consumption of the whole network as well as to reduce the failure rate in each device that may occur due to a high level of link congestion and elevated temperature. The experimental results show that uplink utilization has been reduced by 8.33% and power consumption levels reduced by 0.85kw/day which is equivalent to 34.9% of the operational power. Additionally, the high raised temperatures have been dropped from high range 50+ Cᵒ (high critical) to mid 40+ Cᵒ (high warning). The effectiveness of the proposed approach was verified experimentally using a virtualized testbed platform.
ISSN: 2169-3536
Appears in Collections:Dept of Electronic and Computer Engineering Research Papers

Files in This Item:
File Description SizeFormat 
FullText.pdf6.66 MBAdobe PDFView/Open

Items in BURA are protected by copyright, with all rights reserved, unless otherwise indicated.