Pub Date : 2018-06-01DOI: 10.1109/NETSOFT.2018.8460110
Fabian Kurtz, Caner Bektas, N. Dorsch, C. Wietfeld
Increasing automation in systems such as Smart Grids (SGs), Intelligent Transportation, the Internet of Things (IoT) and Industry 4.0, involves the need for robust, highly capable Information and Communication Technology (ICT). Traditionally, to meet diverging use case requirements regarding network data rate, delay, security, reliability and flexibility, dedicated communication infrastructures are employed. Yet, this is associated with high costs and lengthy roll-out times. Therefore it is desirable for multiple tenants to share one Physical Network (PN). However, this may compromise service guarantees, potentially violating Service Level Agreements (SLAs). Network slicing aims to address this challenge by transparently dividing one common infrastructure into multiple, logically independent networks. Thereby tenants are isolated from one another, ensuring the fulfillment of hard performance guarantees. As slicing is central to realizing the potential of 5G networks, this work presents a novel approach based on Network Function Virtualization (NFV) and Software-Defined Networking (SDN) driven queueing strategies. The developed solution is comprehensively evaluated with realistic traffic in a physical testing environment. Highly demanding critical infrastructure use cases, with multiple service levels per slice, are used to validate performance and demonstrate functionalities such as dynamic data rate allocation.
{"title":"Network Slicing for Critical Communications in Shared 5G Infrastructures - An Empirical Evaluation","authors":"Fabian Kurtz, Caner Bektas, N. Dorsch, C. Wietfeld","doi":"10.1109/NETSOFT.2018.8460110","DOIUrl":"https://doi.org/10.1109/NETSOFT.2018.8460110","url":null,"abstract":"Increasing automation in systems such as Smart Grids (SGs), Intelligent Transportation, the Internet of Things (IoT) and Industry 4.0, involves the need for robust, highly capable Information and Communication Technology (ICT). Traditionally, to meet diverging use case requirements regarding network data rate, delay, security, reliability and flexibility, dedicated communication infrastructures are employed. Yet, this is associated with high costs and lengthy roll-out times. Therefore it is desirable for multiple tenants to share one Physical Network (PN). However, this may compromise service guarantees, potentially violating Service Level Agreements (SLAs). Network slicing aims to address this challenge by transparently dividing one common infrastructure into multiple, logically independent networks. Thereby tenants are isolated from one another, ensuring the fulfillment of hard performance guarantees. As slicing is central to realizing the potential of 5G networks, this work presents a novel approach based on Network Function Virtualization (NFV) and Software-Defined Networking (SDN) driven queueing strategies. The developed solution is comprehensively evaluated with realistic traffic in a physical testing environment. Highly demanding critical infrastructure use cases, with multiple service levels per slice, are used to validate performance and demonstrate functionalities such as dynamic data rate allocation.","PeriodicalId":333377,"journal":{"name":"2018 4th IEEE Conference on Network Softwarization and Workshops (NetSoft)","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128327014","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-06-01DOI: 10.1109/NETSOFT.2018.8460117
Krishna P. Kadiyala, J. Cobb
The growth of Software Defined Networking (SDN) has made it appealing for Internet service providers to incorporate SDN into their existing legacy networks. This has led to the idea of incrementally introducing SDN elements into the existing legacy infrastructure. Transitioning to such a hybrid SDN network is no small task by itself, and requires carefully planning which nodes in the legacy network can take on the role of SDN elements. The focus of this paper is to identify a set of candidate nodes in an Autonomous System (AS) that can be upgraded to SDN nodes with the objective of minimizing the maximum link utilization at the inter-AS links of the service provider. To this end, we first introduce the SDN node selection problem. Due to its intractability, we propose different greedy heuristics to help select the SDN nodes and show that by selecting the right set of candidate SDN nodes, a small fraction of SDN nodes help achieve a significant reduction in link utilization.
{"title":"SDN Node Selection for Inter-AS TE","authors":"Krishna P. Kadiyala, J. Cobb","doi":"10.1109/NETSOFT.2018.8460117","DOIUrl":"https://doi.org/10.1109/NETSOFT.2018.8460117","url":null,"abstract":"The growth of Software Defined Networking (SDN) has made it appealing for Internet service providers to incorporate SDN into their existing legacy networks. This has led to the idea of incrementally introducing SDN elements into the existing legacy infrastructure. Transitioning to such a hybrid SDN network is no small task by itself, and requires carefully planning which nodes in the legacy network can take on the role of SDN elements. The focus of this paper is to identify a set of candidate nodes in an Autonomous System (AS) that can be upgraded to SDN nodes with the objective of minimizing the maximum link utilization at the inter-AS links of the service provider. To this end, we first introduce the SDN node selection problem. Due to its intractability, we propose different greedy heuristics to help select the SDN nodes and show that by selecting the right set of candidate SDN nodes, a small fraction of SDN nodes help achieve a significant reduction in link utilization.","PeriodicalId":333377,"journal":{"name":"2018 4th IEEE Conference on Network Softwarization and Workshops (NetSoft)","volume":"96 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134045508","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-06-01DOI: 10.1109/NETSOFT.2018.8460115
G. Sviridov, M. Bonola, Angelo Tulumello, P. Giaccone, A. Bianco, G. Bianchi
In stateful Software Defined Network (SDN) data planes, network switches hold some local flow-related states thanks to which they are able to perform decisions by locally executing simple algorithms. While stateful data planes provide better reactiveness in respect to vanilla SDN, states are still bounded to single switches which represents a significant restraint for network-wide applications. To deal with the current limitations of stateful data planes we introduce and provide design guidelines for LODGE, a model according to which distributed network applications are able to make local decisions at each switche based on some global variables shared across other switches. We describe the implementation of LODGE with a basic application providing support for the detection of Distributed Denial of Service (DDoS) attack in a scenario of stateful data planes involving P4 and Open Packet Processor (OPP) enabled switches. We validate the two implementations in a small emulated testbed and we show the beneficial effects on the reduction of the total network traffic.
{"title":"LODGE: LOcal Decisions on Global statEs in progrananaable data planes","authors":"G. Sviridov, M. Bonola, Angelo Tulumello, P. Giaccone, A. Bianco, G. Bianchi","doi":"10.1109/NETSOFT.2018.8460115","DOIUrl":"https://doi.org/10.1109/NETSOFT.2018.8460115","url":null,"abstract":"In stateful Software Defined Network (SDN) data planes, network switches hold some local flow-related states thanks to which they are able to perform decisions by locally executing simple algorithms. While stateful data planes provide better reactiveness in respect to vanilla SDN, states are still bounded to single switches which represents a significant restraint for network-wide applications. To deal with the current limitations of stateful data planes we introduce and provide design guidelines for LODGE, a model according to which distributed network applications are able to make local decisions at each switche based on some global variables shared across other switches. We describe the implementation of LODGE with a basic application providing support for the detection of Distributed Denial of Service (DDoS) attack in a scenario of stateful data planes involving P4 and Open Packet Processor (OPP) enabled switches. We validate the two implementations in a small emulated testbed and we show the beneficial effects on the reduction of the total network traffic.","PeriodicalId":333377,"journal":{"name":"2018 4th IEEE Conference on Network Softwarization and Workshops (NetSoft)","volume":"58 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123444694","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-06-01DOI: 10.1109/NETSOFT.2018.8460137
W. S. Atoui, I. B. Yahia, Walid Gaaloul
Software Defined Networks (SDN) are becoming a cornerstone in future network architectures. In this regard, building a global network view upon heterogeneous multiple SDN controllers is a challenging task. In fact, each controller exposes its local network view using different conceptual model. To address this problem, we propose a semantic-based framework that encompasses an approach to extract a local ontology from each controller. The framework incorporates algorithms that map the ontologies together and form the overall network view. We apply our framework in two scenarios: a centralized scenario where the SDN controllers expose their network views to a centralized entity (e.g. management application) that builds the network view, and a distributed scenario, where SDN controllers exchange in peer to peer their local network views and build locally the global view. We present also an evaluation of our framework with respect to performance and accuracy and discuss future works.
{"title":"Semantic-Based Global Network View Construction in Software Defined Networks with Multiple Controllers","authors":"W. S. Atoui, I. B. Yahia, Walid Gaaloul","doi":"10.1109/NETSOFT.2018.8460137","DOIUrl":"https://doi.org/10.1109/NETSOFT.2018.8460137","url":null,"abstract":"Software Defined Networks (SDN) are becoming a cornerstone in future network architectures. In this regard, building a global network view upon heterogeneous multiple SDN controllers is a challenging task. In fact, each controller exposes its local network view using different conceptual model. To address this problem, we propose a semantic-based framework that encompasses an approach to extract a local ontology from each controller. The framework incorporates algorithms that map the ontologies together and form the overall network view. We apply our framework in two scenarios: a centralized scenario where the SDN controllers expose their network views to a centralized entity (e.g. management application) that builds the network view, and a distributed scenario, where SDN controllers exchange in peer to peer their local network views and build locally the global view. We present also an evaluation of our framework with respect to performance and accuracy and discuss future works.","PeriodicalId":333377,"journal":{"name":"2018 4th IEEE Conference on Network Softwarization and Workshops (NetSoft)","volume":"61 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124722482","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-06-01DOI: 10.1109/NETSOFT.2018.8459974
Takuya Tojo, Shingo Okada, Yoshiyuki Hirata, S. Yasukawa
In the 5G era, it is expected that various services will be provided on network slices, which are virtualized network resources. A network slice has to support isolation capabilities and multi-grade performance capabilities such as ultra-low-latency and ultra-reliability due to the 5G requirements. IP and optical networks are also required to provide multi-grade property as a part of a network slice. Multi-layer SDN control is beneficial technology to realize the multi-grade network because an integrated control of IP and optical layers enables to combine network technologies across layers. It brings flexibility of service levels in terms of network performance and reliability. In this paper, we propose and demonstrate a multi-grade network which provides various types of virtualized network according to user requirements. Our platform is composed of IP and optical layers and ONOS-based SDN controller. We show on-demand provisioning of a virtualized network with a provisioning tracker that presents current status of provisioning in real-time.
{"title":"An Integrated Control of IP and Optical Network for Multi-grade Virtualized Networks","authors":"Takuya Tojo, Shingo Okada, Yoshiyuki Hirata, S. Yasukawa","doi":"10.1109/NETSOFT.2018.8459974","DOIUrl":"https://doi.org/10.1109/NETSOFT.2018.8459974","url":null,"abstract":"In the 5G era, it is expected that various services will be provided on network slices, which are virtualized network resources. A network slice has to support isolation capabilities and multi-grade performance capabilities such as ultra-low-latency and ultra-reliability due to the 5G requirements. IP and optical networks are also required to provide multi-grade property as a part of a network slice. Multi-layer SDN control is beneficial technology to realize the multi-grade network because an integrated control of IP and optical layers enables to combine network technologies across layers. It brings flexibility of service levels in terms of network performance and reliability. In this paper, we propose and demonstrate a multi-grade network which provides various types of virtualized network according to user requirements. Our platform is composed of IP and optical layers and ONOS-based SDN controller. We show on-demand provisioning of a virtualized network with a provisioning tracker that presents current status of provisioning in real-time.","PeriodicalId":333377,"journal":{"name":"2018 4th IEEE Conference on Network Softwarization and Workshops (NetSoft)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122674292","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-06-01DOI: 10.1109/NETSOFT.2018.8459960
G. Faraci, A. Raciti, S. Rizzo, G. Schembra
Smart agriculture, representing one of the use cases for Industry 4.0, encounters some deployment problems in wide rural areas where neither telecommunications infrastructure, nor a local power grid is available. For this reason, this paper proposes a comprehensive platform constituted by a Flying Ad hoc NETwork (FANET) to provide connectivity, and a 5G network access node equipped with MEC facilities to run software applications at the edge of the network. Power supply for the 5G base station, the drones to be periodically recharged and the servers in the data center equipped with MEC facilities is realized by an electrical hybrid generator. An analytical model of the whole system is also defined to capture all the above issues, and supporting the design of the proposed system.
{"title":"A 5G platform for Unmanned Aerial Monitoring in Rural Areas: Design and Performance Issues","authors":"G. Faraci, A. Raciti, S. Rizzo, G. Schembra","doi":"10.1109/NETSOFT.2018.8459960","DOIUrl":"https://doi.org/10.1109/NETSOFT.2018.8459960","url":null,"abstract":"Smart agriculture, representing one of the use cases for Industry 4.0, encounters some deployment problems in wide rural areas where neither telecommunications infrastructure, nor a local power grid is available. For this reason, this paper proposes a comprehensive platform constituted by a Flying Ad hoc NETwork (FANET) to provide connectivity, and a 5G network access node equipped with MEC facilities to run software applications at the edge of the network. Power supply for the 5G base station, the drones to be periodically recharged and the servers in the data center equipped with MEC facilities is realized by an electrical hybrid generator. An analytical model of the whole system is also defined to capture all the above issues, and supporting the design of the proposed system.","PeriodicalId":333377,"journal":{"name":"2018 4th IEEE Conference on Network Softwarization and Workshops (NetSoft)","volume":"79 12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115827972","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-06-01DOI: 10.1109/NETSOFT.2018.8459997
D. Harris, J. Naor, D. Raz
Multi-access Edge Computing (MEC) is a new network architecture that allows applications and network services to be executed at the edge of the network. This is done by running these services on commodity servers that are placed in close proximity to the network edge and to the cellular base stations in wireless networks. This architecture provides high bandwidth and low latency for network functions and other applications. However, the availability of the resources at the network edge is limited and thus one of the main challenges in deploying this new paradigm is the ability to locate these latency sensitive services in the appropriate network location according to the specific demand for each service and the relevant latency constraints. In this paper we address this challenge by defining the Virtual Network Functions Placement and Assignment Problem (VNFPAP) and providing algorithms with guaranteed performance for it. We also show by simulating our algorithms on real mobile data that in realistic scenarios they perform much better than current used heuristics.
{"title":"Latency Aware Placement in Multi-access Edge Computing","authors":"D. Harris, J. Naor, D. Raz","doi":"10.1109/NETSOFT.2018.8459997","DOIUrl":"https://doi.org/10.1109/NETSOFT.2018.8459997","url":null,"abstract":"Multi-access Edge Computing (MEC) is a new network architecture that allows applications and network services to be executed at the edge of the network. This is done by running these services on commodity servers that are placed in close proximity to the network edge and to the cellular base stations in wireless networks. This architecture provides high bandwidth and low latency for network functions and other applications. However, the availability of the resources at the network edge is limited and thus one of the main challenges in deploying this new paradigm is the ability to locate these latency sensitive services in the appropriate network location according to the specific demand for each service and the relevant latency constraints. In this paper we address this challenge by defining the Virtual Network Functions Placement and Assignment Problem (VNFPAP) and providing algorithms with guaranteed performance for it. We also show by simulating our algorithms on real mobile data that in realistic scenarios they perform much better than current used heuristics.","PeriodicalId":333377,"journal":{"name":"2018 4th IEEE Conference on Network Softwarization and Workshops (NetSoft)","volume":"122 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123712786","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-06-01DOI: 10.1109/NETSOFT.2018.8460111
Bong-yeol Yu, Gyeongsik Yang, C. Yoo
In SDN, as the control channel becomes a performance bottleneck, modeling the control channel traffic is important. Such a model is useful in predicting the control channel traffic for network provisioning. However, previously proposed models are quite limited in that they assume only the forwarding function of a specific controller for their models. To overcome the limitations, first, this paper analyzes the control traffic by seven functions (including forwarding function) of a controller. Then, we build a seven-function model to predict control channel usage and evaluate the prediction accuracy that achieves as high as 94%. Note that previous models did not have any quantitative evaluation. Our model is built with the Open Network Operating System (ONOS) controller and extended to Floodlight and POX controllers. We show that the extended model also achieves similar prediction accuracy (95%). Furthermore, we compare three controllers in terms of control channel usage through our model.
{"title":"Comprehensive Prediction Models of Control Traffic for SDN Controllers","authors":"Bong-yeol Yu, Gyeongsik Yang, C. Yoo","doi":"10.1109/NETSOFT.2018.8460111","DOIUrl":"https://doi.org/10.1109/NETSOFT.2018.8460111","url":null,"abstract":"In SDN, as the control channel becomes a performance bottleneck, modeling the control channel traffic is important. Such a model is useful in predicting the control channel traffic for network provisioning. However, previously proposed models are quite limited in that they assume only the forwarding function of a specific controller for their models. To overcome the limitations, first, this paper analyzes the control traffic by seven functions (including forwarding function) of a controller. Then, we build a seven-function model to predict control channel usage and evaluate the prediction accuracy that achieves as high as 94%. Note that previous models did not have any quantitative evaluation. Our model is built with the Open Network Operating System (ONOS) controller and extended to Floodlight and POX controllers. We show that the extended model also achieves similar prediction accuracy (95%). Furthermore, we compare three controllers in terms of control channel usage through our model.","PeriodicalId":333377,"journal":{"name":"2018 4th IEEE Conference on Network Softwarization and Workshops (NetSoft)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130696465","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-06-01DOI: 10.1109/NETSOFT.2018.8459946
K. A. Noghani, A. Kassler, Prem Sankar Gopannan
Live Virtual Machine (VM) migration has significantly improved the flexibility of modern Data Centers (DC). However, seamless live migration of a VM between geo-distributed DCs faces several challenges due to difficulties in preserving the network configuration after the migration paired with a large network convergence time. Although SDN-based approaches can speed up network convergence time, these techniques have two limitations. First, they typically react to the new topology by installing new flow rules once the migration is finished. Second, because the WAN is typically not under SDN control, they result in sub-optimal routing thus severely degrading the network performance once the VM is attached at the new location. In this paper, we identify networking challenges for VM migration across geo-distributed DCs. Based on those observations, we design a novel long-haul VM migration scheme that overcomes those limitations. First, instead of reactively restoring connectivity after the migration, our SDN-based approach proactively restores flows across the WAN towards the new location with the help of EVPN and VXLAN overlay technologies. Second, the SDN controller accelerates the network convergence by announcing the migration to other controllers using MP-BGP control plane messages. Finally, the SDN controller resolves the sub-optimal routing problem that arises as a result of migration implementing a distributed anycast gateway. We implement our approach as extensions to the OpenDaylight controller. Our evaluation shows that our approach outperforms existing approaches in reducing the downtime by 400 ms and increasing the application performance up to 12 times.
{"title":"EVPN/SDN Assisted Live VM Migration between Geo-Distributed Data Centers","authors":"K. A. Noghani, A. Kassler, Prem Sankar Gopannan","doi":"10.1109/NETSOFT.2018.8459946","DOIUrl":"https://doi.org/10.1109/NETSOFT.2018.8459946","url":null,"abstract":"Live Virtual Machine (VM) migration has significantly improved the flexibility of modern Data Centers (DC). However, seamless live migration of a VM between geo-distributed DCs faces several challenges due to difficulties in preserving the network configuration after the migration paired with a large network convergence time. Although SDN-based approaches can speed up network convergence time, these techniques have two limitations. First, they typically react to the new topology by installing new flow rules once the migration is finished. Second, because the WAN is typically not under SDN control, they result in sub-optimal routing thus severely degrading the network performance once the VM is attached at the new location. In this paper, we identify networking challenges for VM migration across geo-distributed DCs. Based on those observations, we design a novel long-haul VM migration scheme that overcomes those limitations. First, instead of reactively restoring connectivity after the migration, our SDN-based approach proactively restores flows across the WAN towards the new location with the help of EVPN and VXLAN overlay technologies. Second, the SDN controller accelerates the network convergence by announcing the migration to other controllers using MP-BGP control plane messages. Finally, the SDN controller resolves the sub-optimal routing problem that arises as a result of migration implementing a distributed anycast gateway. We implement our approach as extensions to the OpenDaylight controller. Our evaluation shows that our approach outperforms existing approaches in reducing the downtime by 400 ms and increasing the application performance up to 12 times.","PeriodicalId":333377,"journal":{"name":"2018 4th IEEE Conference on Network Softwarization and Workshops (NetSoft)","volume":"76 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121033218","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-06-01DOI: 10.1109/NETSOFT.2018.8459979
Angelos Mimidis Kentis, Artur Pilimon, José Soler, M. Berger, S. Ruepp
The forwarding rules, used by the legacy and SDN network devices to perform routing/forwarding decisions, are generally stored in Ternary Content Addressable Memory (TCAM) modules, which offer constant look-up times, but have limited capacity, due to their high capital and operational costs, high power consumption and high silicon footprint. To counter this limitation, some commercial switches offer both, hardware and software flow table implementations, termed hybrid flow table architecture in this paper. The software-based tables are stored in non-TCAM memory modules, which offer higher capacity, but with slower lookup times. In addition, these memory modules are limited in terms of how many requests they can serve per time unit. Thus, exceeding this threshold will lead to packet loss in the network. This paper proposes a novel placement algorithm, which dynamically decides whether a new flow rule should be placed in a hardware (expensive) or a software (cheap) table. The placement decisions are based on a number of criteria with the goal to increase the utilization of the software-based table, without introducing performance degradation in the network in terms of significant delay and packet loss. The performance of the placement algorithm was evaluated through experimental measurements in a testbed, which comprises a hybrid SDN switch, a server performing traffic generation and a server hosting the SDN controller. The results indicate that, by limiting the maximum allowed processing capacity of the software table, the number of accommodated flows is significantly increased, while bounding any excessive delays and avoiding packet loss.
{"title":"A Novel Algorithm for Flow-Rule Placement in SDN Switches","authors":"Angelos Mimidis Kentis, Artur Pilimon, José Soler, M. Berger, S. Ruepp","doi":"10.1109/NETSOFT.2018.8459979","DOIUrl":"https://doi.org/10.1109/NETSOFT.2018.8459979","url":null,"abstract":"The forwarding rules, used by the legacy and SDN network devices to perform routing/forwarding decisions, are generally stored in Ternary Content Addressable Memory (TCAM) modules, which offer constant look-up times, but have limited capacity, due to their high capital and operational costs, high power consumption and high silicon footprint. To counter this limitation, some commercial switches offer both, hardware and software flow table implementations, termed hybrid flow table architecture in this paper. The software-based tables are stored in non-TCAM memory modules, which offer higher capacity, but with slower lookup times. In addition, these memory modules are limited in terms of how many requests they can serve per time unit. Thus, exceeding this threshold will lead to packet loss in the network. This paper proposes a novel placement algorithm, which dynamically decides whether a new flow rule should be placed in a hardware (expensive) or a software (cheap) table. The placement decisions are based on a number of criteria with the goal to increase the utilization of the software-based table, without introducing performance degradation in the network in terms of significant delay and packet loss. The performance of the placement algorithm was evaluated through experimental measurements in a testbed, which comprises a hybrid SDN switch, a server performing traffic generation and a server hosting the SDN controller. The results indicate that, by limiting the maximum allowed processing capacity of the software table, the number of accommodated flows is significantly increased, while bounding any excessive delays and avoiding packet loss.","PeriodicalId":333377,"journal":{"name":"2018 4th IEEE Conference on Network Softwarization and Workshops (NetSoft)","volume":"50 4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125202323","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：