How physical network topologies affect virtual network embedding quality: A characterization study based on ISP and datacenter networks

Abstract Network virtualization is a mechanism that allows the coexistence of multiple virtual networks on top of a single physical substrate. Due to its well-known potential benefits (e.g., lower CAPEX/OPEX expenditures), it has been embraced by the IT sector, specially by Internet Service Providers (ISPs) and cloud computing/datacenter companies. One of the research challenges addressed recently in the literature is the efficient mapping of virtual resources on physical infrastructures. Although this challenge has received considerable attention, state-of-the-art approaches present, in general, a high rejection rate, i.e., the ratio between the number of denied virtual network requests and the total amount of requests is considerably high. In this work, we investigate the relationship between the quality of virtual network mappings and the topological structures of the underlying substrates. Exact solutions of an online embedding model are evaluated under different classes of ISP and datacenter network topologies. The obtained results demonstrate that the employment of physical topologies that contain regions with high connectivity significantly contributes to the reduction of rejection rates and, therefore, to improved resource usage. Additionally, through an extensive analysis of denied requests, we assess the main rejection causes related to both ISP and datacenter networks and provide strong evidence of each one. In summary, through the embedding of virtual requests, available resources in ISP networks tend to be more partitioned in comparison to datacenter networks. Such differences on partitioning levels lead to a different percentage of rejection causes in each topology class.

[1]  Raouf Boutaba,et al.  ViNEYard: Virtual Network Embedding Algorithms With Coordinated Node and Link Mapping , 2012, IEEE/ACM Transactions on Networking.

[2]  Xiang Cheng,et al.  Adaptive multi-objective artificial immune system based virtual network embedding , 2015, J. Netw. Comput. Appl..

[3]  Alejandro López-Ortiz,et al.  REWIRE: An optimization-based framework for unstructured data center network design , 2012, 2012 Proceedings IEEE INFOCOM.

[4]  Tatsuya Mori,et al.  Impact of topology on parallel video streaming , 2010, 2010 IEEE Network Operations and Management Symposium - NOMS 2010.

[5]  Xiang Cheng,et al.  Virtual network embedding through topology awareness and optimization , 2012, Comput. Networks.

[6]  Djamal Zeghlache,et al.  Virtual network provisioning across multiple substrate networks , 2011, Comput. Networks.

[7]  Minlan Yu,et al.  Rethinking virtual network embedding: substrate support for path splitting and migration , 2008, CCRV.

[8]  Amin Vahdat,et al.  Data Center Switch Architecture in the Age of Merchant Silicon , 2009, 2009 17th IEEE Symposium on High Performance Interconnects.

[9]  Helen J. Wang,et al.  SecondNet: a data center network virtualization architecture with bandwidth guarantees , 2010, CoNEXT.

[10]  RexfordJennifer,et al.  Rethinking virtual network embedding , 2008 .

[11]  Deep Medhi,et al.  Opportunistic resilience embedding (ORE): Toward cost-efficient resilient virtual networks , 2015, Comput. Networks.

[12]  Luciana S. Buriol,et al.  Characterizing the impact of network substrate topologies on virtual network embedding , 2013, Proceedings of the 9th International Conference on Network and Service Management (CNSM 2013).

[13]  Charles E. Leiserson,et al.  Fat-trees: Universal networks for hardware-efficient supercomputing , 1985, IEEE Transactions on Computers.

[14]  Albert G. Greenberg,et al.  Sharing the Data Center Network , 2011, NSDI.

[15]  Mohamed Faten Zhani,et al.  Venice: Reliable virtual data center embedding in clouds , 2014, IEEE INFOCOM 2014 - IEEE Conference on Computer Communications.

[16]  David A. Maltz,et al.  Network traffic characteristics of data centers in the wild , 2010, IMC '10.

[17]  Xiang Cheng,et al.  Virtual network embedding through topology-aware node ranking , 2011, CCRV.

[18]  Hermann de Meer,et al.  Distributed and scalable embedding of virtual networks , 2015, J. Netw. Comput. Appl..

[19]  Miguel Rio,et al.  Network topologies: inference, modeling, and generation , 2008, IEEE Communications Surveys & Tutorials.

[20]  Albert G. Greenberg,et al.  VL2: a scalable and flexible data center network , 2009, SIGCOMM '09.

[21]  Lisandro Zambenedetti Granville,et al.  Data Center Network Virtualization: A Survey , 2013, IEEE Communications Surveys & Tutorials.

[22]  Hitesh Ballani,et al.  Towards predictable datacenter networks , 2011, SIGCOMM 2011.

[23]  Gwendal Simon,et al.  VDC Planner: Dynamic migration-aware Virtual Data Center embedding for clouds , 2013, 2013 IFIP/IEEE International Symposium on Integrated Network Management (IM 2013).

[24]  Guy Pujolle,et al.  VNE-AC: Virtual Network Embedding Algorithm Based on Ant Colony Metaheuristic , 2011, 2011 IEEE International Conference on Communications (ICC).

[25]  Luciana S. Buriol,et al.  Trust-based grouping for cloud datacenters: Improving security in shared infrastructures , 2013, 2013 IFIP Networking Conference.

[26]  Haitao Wu,et al.  BCube: a high performance, server-centric network architecture for modular data centers , 2009, SIGCOMM '09.

[27]  Luciana S. Buriol,et al.  A heuristic-based algorithm for privacy-oriented virtual network embedding , 2014, 2014 IEEE Network Operations and Management Symposium (NOMS).

[28]  Albert,et al.  Topology of evolving networks: local events and universality , 2000, Physical review letters.

[29]  Gustavo Prado Alkmim,et al.  Mapping virtual networks onto substrate networks , 2013, Journal of Internet Services and Applications.

[30]  Edmundo Roberto Mauro Madeira,et al.  Virtual network security: threats, countermeasures, and challenges , 2015, Journal of Internet Services and Applications.