Service function chain network planning through offline, online and infeasibility restoration techniques
Ramy Mohamed,Marios Avgeris,Aris Leivadeas,Ioannis Lambadaris,John Chinneck,Todd Morris,Petar Djukic
DOI: https://doi.org/10.1016/j.comnet.2024.110241
IF: 5.493
2024-04-01
Computer Networks
Abstract:Network Function Virtualization (NFV) and Service Function Chaining (SFC) have created a new, flexible, and cost-efficient way of network service offering. Thus, Service (SPs) and Infrastructure Providers (InPs) are updating their product portfolio to incorporate these new technologies. However, placing the network components, called Virtualized Network Functions (VNFs), that constitute the network services on top of the physical infrastructure creates several resource allocation challenges. This paper offers a networking planning solution tool for SP and InPs to allocate resources according to the expected network demand appropriately. In particular, first, we propose an offline and optimal solution that dimensions and exactly allocates the physical resources according to the estimated workload. Following, mispredicted offline solutions are corrected by various online placement solutions that serve the network workload according to the time of its arrival. Finally, to alleviate any resource mismatch between the total capacity of the physical network and the total requested resources of the expected SFCs, a novel feasibility placement approach is proposed that provides suggestions to the InPs on where to place additional resources. Extensive experimentation shows that the proposed planning tool can efficiently preallocate SFCs under different configuration scenarios, while the feasibility restoration tool provides accurate, timely, and cost-efficient remedy solutions for the InPs.
computer science, information systems,telecommunications,engineering, electrical & electronic, hardware & architecture