%0 Conference Paper %1 Blenk2018InfocomFull %A Blenk, Andreas %A Kalmbach, Patrick %A Zerwas, Johannes %A Jarschel, Michael %A Schmid, Stefan %A Kellerer, Wolfgang %B 37th IEEE Conference on Computer Communications (INFOCOM) %D 2018 %I IEEE %K SENDATE SENDATE-NETWORKING SENDATE-PLANETS myown %P 405-413 %T NeuroViNE: A Neural Preprocessor for Your Virtual Network Embedding Algorithm %U http://dblp.uni-trier.de/db/conf/infocom/infocom2018.html#BlenkKZJ0K18 %X Network virtualization enables the increasingly diverse network services to cohabit and share a given physical infrastructure and its resources, while potentially relying on different network architectures and protocols optimized towards the specific requirements. However, in order to ensure a predictable performance despite the shared resources, network virtualization requires a strict performance isolation and hence resource reservations. Moreover, the creation of virtual networks should be fast and efficient. The underlying NP-hard algorithmic problem is known as the Virtual Network Embedding (VNE) problem and has been studied intensively over the last years. This paper presents NeuroViNE, a novel approach to speed up and improve a wide range of existing VNE algorithms: NeuroViNE is based on a search space reduction mechanism and preprocesses a problem instance by extracting relevant subgraphs, i.e., good combinations of substrate nodes and links. These subgraphs can then be fed to an existing algorithm for faster and more resource-efficient placements. NeuroViNE relies on a Hopfield network, and its performance benefits are investigated in extensive simulations for random networks, real substrate networks, and data center networks.

@inproceedings{Blenk2018InfocomFull, abstract = {Network virtualization enables the increasingly diverse network services to cohabit and share a given physical infrastructure and its resources, while potentially relying on different network architectures and protocols optimized towards the specific requirements. However, in order to ensure a predictable performance despite the shared resources, network virtualization requires a strict performance isolation and hence resource reservations. Moreover, the creation of virtual networks should be fast and efficient. The underlying NP-hard algorithmic problem is known as the Virtual Network Embedding (VNE) problem and has been studied intensively over the last years. This paper presents NeuroViNE, a novel approach to speed up and improve a wide range of existing VNE algorithms: NeuroViNE is based on a search space reduction mechanism and preprocesses a problem instance by extracting relevant subgraphs, i.e., good combinations of substrate nodes and links. These subgraphs can then be fed to an existing algorithm for faster and more resource-efficient placements. NeuroViNE relies on a Hopfield network, and its performance benefits are investigated in extensive simulations for random networks, real substrate networks, and data center networks.}, added-at = {2019-01-24T12:19:19.000+0100}, author = {Blenk, Andreas and Kalmbach, Patrick and Zerwas, Johannes and Jarschel, Michael and Schmid, Stefan and Kellerer, Wolfgang}, biburl = {https://www.bibsonomy.org/bibtex/2594eb3303aa26e0f48d6d2144722fcb8/andreasblenk}, booktitle = {37th IEEE Conference on Computer Communications (INFOCOM)}, interhash = {94f6a117521799985842abfbf20073ee}, intrahash = {594eb3303aa26e0f48d6d2144722fcb8}, keywords = {SENDATE SENDATE-NETWORKING SENDATE-PLANETS myown}, month = apr, pages = {405-413}, publisher = {IEEE}, timestamp = {2019-01-24T12:19:19.000+0100}, title = {NeuroViNE: A Neural Preprocessor for Your Virtual Network Embedding Algorithm}, url = {http://dblp.uni-trier.de/db/conf/infocom/infocom2018.html#BlenkKZJ0K18}, year = 2018 }