TY - GEN
T1 - An Energy Efficient SDN Controller Placement with Delay Constraints
AU - Kondo, Tomofumi
AU - Guillen, Luis
AU - Izumi, Satoru
AU - Abe, Toru
AU - Mizuki, Takaaki
AU - Suganuma, Takuo
N1 - Publisher Copyright:
Copyright 2023 KICS.
PY - 2023
Y1 - 2023
N2 - Software-Defined Networking (SDN) separates the control from the data plane in the network infrastructure to flexibly and efficiently manage the resources. Since the controller is the core of the control plane, multiple instances need to be placed within the network to increase performance, reliability, and fault tolerance, among others. However, deciding the number and where to place the allotted controllers is not trivial; it is known as the Controller Placement Problem (CPP). This paper considers the energy-efficiency aspect of the network infrastructure to solve the CPP. To do so, we designed a model that includes delay constraints, whose effectivity was assessed and confirmed to save about 20-30% more energy than a model that only considers the number of controllers. Moreover, it is confirmed that execution time can be decreased by 100 times while keeping the energy consumption low by reducing the number of pre-calculated paths between the SDN switches and the controllers.
AB - Software-Defined Networking (SDN) separates the control from the data plane in the network infrastructure to flexibly and efficiently manage the resources. Since the controller is the core of the control plane, multiple instances need to be placed within the network to increase performance, reliability, and fault tolerance, among others. However, deciding the number and where to place the allotted controllers is not trivial; it is known as the Controller Placement Problem (CPP). This paper considers the energy-efficiency aspect of the network infrastructure to solve the CPP. To do so, we designed a model that includes delay constraints, whose effectivity was assessed and confirmed to save about 20-30% more energy than a model that only considers the number of controllers. Moreover, it is confirmed that execution time can be decreased by 100 times while keeping the energy consumption low by reducing the number of pre-calculated paths between the SDN switches and the controllers.
KW - BIP
KW - Controller Placement Problem
KW - Energy Efficiency
KW - SDN
UR - http://www.scopus.com/inward/record.url?scp=85174814971&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85174814971&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:85174814971
T3 - APNOMS 2023 - 24th Asia-Pacific Network Operations and Management Symposium: Intelligent Management for Enabling the Digital Transformation
SP - 119
EP - 124
BT - APNOMS 2023 - 24th Asia-Pacific Network Operations and Management Symposium
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 24th Asia-Pacific Network Operations and Management Symposium, APNOMS 2023
Y2 - 6 September 2023 through 8 September 2023
ER -
