Scalable Cost-Efficient Placement and Chaining of Virtual Network Functions
Resumo
While Network Function Virtualization (NFV) is increasingly gaining momentum, with promising benefits of flexible service function deployment and reduced operations and management costs, there are several challenges that remain to be properly tackled, so that it can realize its full potential. One of these challenges, which has a significant impact on the NFV production chain, is effectively and (cost) efficiently deploying service functions, while ensuring that service level agreements are satisfied and making wise allocations of network resources. Despite recent research activity in the field, little has been done towards scalable and cost-efficient placement & chaining of virtual network functions (VNFs) – a key feature for the effective success of NFV. In this thesis, we approach VNF placement and chaining as an optimization problem in the context of Interand Intra-datacenter. We formalize the Virtual Network Function Placement and Chaining (VNFPC) problem and propose a mathematical model to solve it. Our model has established one of the first baseline comparison in the field of resource management in NFV and has been widely used in the recent literature. We also address scalability of VNFPC problem to solve large instances by proposing a novel fix-and-optimize-based heuristic algorithm for tackling it. Further, we extensively measure the performance limitations of realistic NFV deployments. Based on that, we propose an analytical model that accurately predict incurred operational costs. Then, we develop an optimal Intra-datacenter service chain deployment mechanism based on our cost model. Finally, we tackle the problem of monitoring service chains in NFV-based environments efficiently.
Referências
Barkai, S., Katz, R., Farinacci, D., and Meyer, D. (2013). Software defined flow-mapping for scaling virtualized network functions. In Proceedings..., ACM SIGCOMM HotSDN’13, pages 149–150, New York, NY, USA. ACM SIGCOMM Workshop on Hot Topics in Software Defined Networking, ACM.
Group, N. F. I. S. (2012). Network function virtualisation (nfv): An introduction, benefits, enablers, challenges and call for action. Available at: https://portal.etsi.org/nfv/. Accessed on: Jan. 20, 2015.
Lewin-Eytan, L., Naor, J., Cohen, R., and Raz, D. (2015). Near optimal placement of virtual network functions. In Proceedings..., INFOCOM’15, pages 1346–1354, New York, NY, USA. IEEE INFOCOM, IEEE.
Luizelli, M. C., Bays, L. R., Buriol, L. S., Barcellos, M. P., and Gaspary, L. P. (2015). Piecing together the nfv provisioning puzzle: Efficient placement and chaining of virtual network functions. In Proceedings..., IM’15, pages 98–106, New York, NY, USA. IFIP/IEEE International Symposium on Integrated Network Management, IEEE.
Luizelli, M. C., da Costa Cordeiro, W. L., Buriol, L. S., and Gaspary, L. P. (2017a). A fix-andoptimize approach for efficient and large scale virtual network function placement and chaining. Computer Communications, 102:67 – 77.
Luizelli, M. C., Raz, D., Saar, Y., and Yallouz, J. (2017b). The actual cost of software switching for nfv chaining. In Proceedings..., IM’17, pages 335–343, New York, NY, USA. IFIP/IEEE International Symposium on Integrated Network Management, IEEE.
Luizelli, M. C., Saar, Y., and Raz, D. (2018). Optimizing nfv chain deployment through minimizing the cost of virtual switching. In Proceedings of IEEE International Conference on Computer Communications, INFOCOM ’18, pages 1–9, Piscataway, NJ, USA. IEEE INFOCOM, IEEE Press.
Martins, J., Ahmed, M., Raiciu, C., Olteanu, V., Honda, M., Bifulco, R., and Huici, F. (2014). Clickos and the art of network function virtualization. In Proceedings..., NSDI’14, Berkeley, CA, USA. USENIX Conference on Networked Systems Design and Implementation, USENIX Association.
Mechtri, M., Ghribi, C., Soualah, O., and Zeghlache, D. (2017). Nfv orchestration framework addressing sfc challenges. IEEE Communications Magazine, 55(6):16–23.
Mehraghdam, S., Keller, M., and Karl, H. (2014). Specifying and placing chains of virtual network functions. In Proceedings..., CloudNet’14, pages 7–13, Piscataway, NJ, USA. IEEE International Conference on Cloud Networking, IEEE Press.
Moens, H. and De Turck, F. (2014). Vnf-p: A model for efficient placement of virtualized network functions. In Proceedings..., CNSM’14 MiniConf, pages 418–423, Piscataway, NJ, USA. International Conference on Network and Service Management, IEEE Press.
