TY - GEN
T1 - Live migration for service function chaining
AU - Zhao, Dongcheng
AU - Sun, Gang
AU - Liao, Dan
AU - Iqbal, Rahat
AU - Chang, Victor
N1 - Publisher Copyright:
Copyright © 2017 by SCITEPRESS - Science and Technology Publications, Lda. All rights reserved.
PY - 2017
Y1 - 2017
N2 - Network Function Virtualization (NFV) has been proposed to solve these challenges of hardware middle boxes such as high Capital Expenditures (CAPEX) and Operational Expenditures (OPEX). NFV aims to move packet processing from hardware middle boxes to software middle boxes running on commodity hardware. In NVF, users or virtual machines (VMs) communicate through the service function chaining. Therefore, when VMs are migrated, the service function chaining also needs to be migrated. Most research on migration focus on the issue of VM migration, and at present there is little research on the migration problem of the service function chaining. Therefore, in this paper we focus on the service function chaining migration, we will introduce the serial migration strategy and the parallel migration strategy for multiple VMs into the migration problem of the service function chaining, and propose an improved serial migration strategy for the service function chaining that is based on the serial migration strategy. We then present the m mixed migration strategy for the service function chaining that is based on the improved serial migration strategy and the parallel migration strategy. We conduct detailed simulations to evaluate the performance of the m mixed migration strategy in terms of the migration time and the downtime. We also develop the M/M/C/C and the M/M/C queuing models to calculate performance indicators, such as the blocking rate of each migration request.
AB - Network Function Virtualization (NFV) has been proposed to solve these challenges of hardware middle boxes such as high Capital Expenditures (CAPEX) and Operational Expenditures (OPEX). NFV aims to move packet processing from hardware middle boxes to software middle boxes running on commodity hardware. In NVF, users or virtual machines (VMs) communicate through the service function chaining. Therefore, when VMs are migrated, the service function chaining also needs to be migrated. Most research on migration focus on the issue of VM migration, and at present there is little research on the migration problem of the service function chaining. Therefore, in this paper we focus on the service function chaining migration, we will introduce the serial migration strategy and the parallel migration strategy for multiple VMs into the migration problem of the service function chaining, and propose an improved serial migration strategy for the service function chaining that is based on the serial migration strategy. We then present the m mixed migration strategy for the service function chaining that is based on the improved serial migration strategy and the parallel migration strategy. We conduct detailed simulations to evaluate the performance of the m mixed migration strategy in terms of the migration time and the downtime. We also develop the M/M/C/C and the M/M/C queuing models to calculate performance indicators, such as the blocking rate of each migration request.
KW - Middle boxes
KW - Migration
KW - Network function virtualization
KW - Service function chaining
KW - Virtual network function
UR - http://www.scopus.com/inward/record.url?scp=85024404908&partnerID=8YFLogxK
U2 - 10.5220/0006364701490156
DO - 10.5220/0006364701490156
M3 - Conference Proceeding
AN - SCOPUS:85024404908
T3 - IoTBDS 2017 - Proceedings of the 2nd International Conference on Internet of Things, Big Data and Security
SP - 149
EP - 156
BT - IoTBDS 2017 - Proceedings of the 2nd International Conference on Internet of Things, Big Data and Security
A2 - Ramachandran, Muthu
A2 - Munoz, Victor Mendez
A2 - Kantere, Verena
A2 - Wills, Gary
A2 - Walters, Robert
A2 - Chang, Victor
PB - SciTePress
T2 - 2nd International Conference on Internet of Things, Big Data and Security, IoTBDS 2017
Y2 - 24 April 2017 through 26 April 2017
ER -