Reducing Service Deployment Cost Through VNF Sharing

Abstract

Thanks to its computational and forwarding capabilities, the mobile network infrastructure can support several third-party ("vertical") services, each composed of a graph of virtual (network) functions (VNFs). Importantly, one or more VNFs are often common to multiple services, thus the services deployment cost could be reduced by letting the services share the same VNF instance instead of devoting a separate instance to each service. By doing that, however, it is critical that the target KPI (key performance indicators) of all services are met. To this end, we study the VNF sharing problem and make decisions on (i) when sharing VNFs among multiple services is possible, (ii) how to adapt the virtual machines running the shared VNFs to the combined load of the assigned services, and (iii) how to prioritize the services traffic within shared VNFs. All decisions aim to minimize the cost for the mobile operator, subject to requirements on end-to-end service performance, e.g., total delay. Notably, we show that the aforementioned priorities should be managed dynamically and vary across VNFs. We then propose the FlexShare algorithm to provide near-optimal VNF-sharing and priority assignment decisions in polynomial time. We prove that FlexShare is within a constant factor from the optimum and, using real-world VNF graphs, we show that it consistently outperforms baseline solutions.

Dynamic VNF Placement, Resource Allocation and Traffic Routing in 5G

“…An example can be found in [2], which models VMs as M/M/1 PS queues, and proposes a MILP and a heuristic solution to minimize the average service delay, while meeting the constraints on the links and host capacities. The works in [21] and [22] aim instead to minimize, respectively, the operational cost and the energy consumption of VMs and links while ensuring end-toend delay KPI. [22] also allows for VNF sharing and studies the impact of applying priorities to different services within a shared VNF.…”

“…The works in [21] and [22] aim instead to minimize, respectively, the operational cost and the energy consumption of VMs and links while ensuring end-toend delay KPI. [22] also allows for VNF sharing and studies the impact of applying priorities to different services within a shared VNF. Zhang et al [23] tackle the request admission problem to maximize the total throughput, neglecting instead queuing delay at VMs.…”

Sharing Distributed and Heterogeneous Resources toward End-to-End 5G Networks: A Comprehensive Survey and a Taxonomy

“…As VNFs are today base components of network services, it is not unusual that they are common for various network services in parallel. Therefore, Malandrino et al [128] study the opportunities of VNF sharing by considering multiple criteria, such as: i) conditions upon which VNFs can be shared, ii) distribution of the workload per virtual machines that run shared VNFs, and iii) possibilities to prioritize service traffic within shared VNFs. Thus, authors propose FlexShare optimization algorithm for VNF sharing, and show that this algorithm outperforms baseline solutions in terms of achieved KPIs such as service deployment cost, and total delay [128].…”

I-Health: Leveraging Edge Computing and Blockchain for Epidemic Management

“…In this paper, buffer overflows are negligible since it is assumed that N i=1 λ i < µ, where µ is the service rate at the BM. By adopting the well-established M/M/1 queuing model [14] (and the references therein) for the received transactions with equal priorities, the average sojourn time of entity i is defined as…”

“…Thus, for validator i to participate in the verification process, it should receive a cost c i that at least covers its payment to the CFP. This condition is represented in constraint (14), where ρ i represents the payment from validator i to the CFP, in order to acquire the needed resources for the verification process.…”