TY - GEN
T1 - Dynamic virtualized network function allocation in a multi-cloud environment
AU - Leivadeas, Aris
AU - Falkner, Matthias
AU - Lambadaris, Ioannis
AU - Kesidis, George
N1 - Publisher Copyright:
© 2016 IEEE.
PY - 2016/6/27
Y1 - 2016/6/27
N2 - Network Function Virtualization (NFV), Cloud Computing, and Software Defined Networking (SDN) are closely related technologies and their combination can provide complete and high quality services to the end users. Multiple Virtualized Network Functions (VNFs) can be implemented as Virtual Machines (VMs) on top of cloud servers, while the traffic flow traversing these VNFs can be facilitated through SDN creating a service chain paradigm. This model promises high performance attracting Service Providers to offer high quality services while reducing the total capital and operating expenses. In this paper, we study the problem of allocating service chains in a multi-cloud environment, where a random number of users are associated with each service chain creating dynamic flow requirements. The framework consists of allocating service chain instances in all the available cloud providers, while associating the incoming users to an appropriate cloud provider based on a number of criteria related to distance and network resource utilization. Towards this extent, a number of algorithms are proposed and evaluated in terms of their overall performance and fairness achieved.
AB - Network Function Virtualization (NFV), Cloud Computing, and Software Defined Networking (SDN) are closely related technologies and their combination can provide complete and high quality services to the end users. Multiple Virtualized Network Functions (VNFs) can be implemented as Virtual Machines (VMs) on top of cloud servers, while the traffic flow traversing these VNFs can be facilitated through SDN creating a service chain paradigm. This model promises high performance attracting Service Providers to offer high quality services while reducing the total capital and operating expenses. In this paper, we study the problem of allocating service chains in a multi-cloud environment, where a random number of users are associated with each service chain creating dynamic flow requirements. The framework consists of allocating service chain instances in all the available cloud providers, while associating the incoming users to an appropriate cloud provider based on a number of criteria related to distance and network resource utilization. Towards this extent, a number of algorithms are proposed and evaluated in terms of their overall performance and fairness achieved.
UR - http://www.scopus.com/inward/record.url?scp=84979302831&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84979302831&partnerID=8YFLogxK
U2 - 10.1109/ICT.2016.7500394
DO - 10.1109/ICT.2016.7500394
M3 - Conference contribution
AN - SCOPUS:84979302831
T3 - 2016 23rd International Conference on Telecommunications, ICT 2016
BT - 2016 23rd International Conference on Telecommunications, ICT 2016
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 23rd International Conference on Telecommunications, ICT 2016
Y2 - 16 May 2016 through 18 May 2016
ER -