R. Bifulco, T. Dietz, Felipe Huici, Mohamed Ahmed, J. Martins, S. Niccolini, H. Kolbe
Broadband Remote Access Servers (BRASes) play a crucial role in today's networks, handling all traffic coming from access networks (e.g., DSL traffic), applying operator policies and providing the first IP point in the network. It is perhaps unsurprising then, that these are expensive, proprietary, difficult-to-upgrade boxes. They also represent a large, single point of failure, making operators even more reticent to deploy new functionality for fear it might seriously disrupt day-to-day operations. In order to remove some of these barriers to innovation, we advocate for turning BRASes from the monolithic hardware boxes they are today into flexible, virtualized, software-based devices running on inexpensive commodity hardware. As a proofof-concept, we present the implementation and performance of a software BRAS based on ClickOS, a tiny Xen virtual machine designed specifically for network processing. Our software BRAS prototype can establish subscriber sessions at rates above 1, 000 per second, requires only 1MB of memory per 1, 000 established sessions, can boot in milliseconds, and can handle traffic at 10Gb/s for almost all packet sizes.
{"title":"Rethinking Access Networks with High Performance Virtual Software BRASes","authors":"R. Bifulco, T. Dietz, Felipe Huici, Mohamed Ahmed, J. Martins, S. Niccolini, H. Kolbe","doi":"10.1109/EWSDN.2013.8","DOIUrl":"https://doi.org/10.1109/EWSDN.2013.8","url":null,"abstract":"Broadband Remote Access Servers (BRASes) play a crucial role in today's networks, handling all traffic coming from access networks (e.g., DSL traffic), applying operator policies and providing the first IP point in the network. It is perhaps unsurprising then, that these are expensive, proprietary, difficult-to-upgrade boxes. They also represent a large, single point of failure, making operators even more reticent to deploy new functionality for fear it might seriously disrupt day-to-day operations. In order to remove some of these barriers to innovation, we advocate for turning BRASes from the monolithic hardware boxes they are today into flexible, virtualized, software-based devices running on inexpensive commodity hardware. As a proofof-concept, we present the implementation and performance of a software BRAS based on ClickOS, a tiny Xen virtual machine designed specifically for network processing. Our software BRAS prototype can establish subscriber sessions at rates above 1, 000 per second, requires only 1MB of memory per 1, 000 established sessions, can boot in milliseconds, and can handle traffic at 10Gb/s for almost all packet sizes.","PeriodicalId":256710,"journal":{"name":"2013 Second European Workshop on Software Defined Networks","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128325330","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}
We describe the design and implementation of a system for performing Virtual Aggregation, a method for dividing and spreading a forwarding table over multiple forwarding elements, in an SDN/OpenFlow network. The primary use-case for this method is to alleviate the scalability problems caused by a rapidly growing routing table in the Default-Free Zone (i.e. the BGP routers on the Internet), secondly it could be extended to support OpenFlow network abstractions. The design provides a scalable system for quickly dividing and distributing a forwarding table in an SDN environment running NOX and OpenFlow 1.1. After evaluating the system we conclude that our design is fast, scalable, and extensible.
{"title":"Virtual Aggregation Using SDN","authors":"Pontus Sköldström, B. Sanchez","doi":"10.1109/EWSDN.2013.16","DOIUrl":"https://doi.org/10.1109/EWSDN.2013.16","url":null,"abstract":"We describe the design and implementation of a system for performing Virtual Aggregation, a method for dividing and spreading a forwarding table over multiple forwarding elements, in an SDN/OpenFlow network. The primary use-case for this method is to alleviate the scalability problems caused by a rapidly growing routing table in the Default-Free Zone (i.e. the BGP routers on the Internet), secondly it could be extended to support OpenFlow network abstractions. The design provides a scalable system for quickly dividing and distributing a forwarding table in an SDN environment running NOX and OpenFlow 1.1. After evaluating the system we conclude that our design is fast, scalable, and extensible.","PeriodicalId":256710,"journal":{"name":"2013 Second European Workshop on Software Defined Networks","volume":"771 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115754532","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}
Since the introduction of software-defined networking (SDN), scalability has been a major concern. There are different approaches to address this issue, and most of them can be addressed without losing the benefits of SDN. SDN provides a level of flexibility that can accommodate network programming and management at scale. In this work we present the recent approaches, which are proposed to address scalability issue of SDN deployment. We particularly select a hierarchical approach for our performance evaluation study. A mathematical framework based on network calculus is presented and the performance of the selected scalable SDN deployment in terms of upper bound of event processing and buffer sizing of the root SDN controller is reported.
{"title":"Performance Evaluation of a Scalable Software-Defined Networking Deployment","authors":"S. Azodolmolky, P. Wieder, R. Yahyapour","doi":"10.1109/EWSDN.2013.18","DOIUrl":"https://doi.org/10.1109/EWSDN.2013.18","url":null,"abstract":"Since the introduction of software-defined networking (SDN), scalability has been a major concern. There are different approaches to address this issue, and most of them can be addressed without losing the benefits of SDN. SDN provides a level of flexibility that can accommodate network programming and management at scale. In this work we present the recent approaches, which are proposed to address scalability issue of SDN deployment. We particularly select a hierarchical approach for our performance evaluation study. A mathematical framework based on network calculus is presented and the performance of the selected scalable SDN deployment in terms of upper bound of event processing and buffer sizing of the root SDN controller is reported.","PeriodicalId":256710,"journal":{"name":"2013 Second European Workshop on Software Defined Networks","volume":"78 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128326234","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}
M. Scharf, V. Gurbani, T. Voith, M. Stein, W. D. Roome, Greg Soprovich, V. Hilt
Virtual Private Networks (VPNs) are a key component of cloud computing systems, since they provide isolated connectivity between geographically separated users. The elasticity in cloud computing and new usage patterns such as cloud bursting require VPNs to be more dynamic than traditional solutions used by network service providers. Managing and optimizing the topology of VPNs requires insight into the underlying wide area network topology and benefits from new network interfaces currently discussed for Software Defined Networks (SDN). This paper presents the use of the Application-Layer Traffic Optimization (ALTO) protocol for VPN optimization. ALTO is a standardized solution for exposure of abstract topology information to a variety of applications, including cloud management systems. We demonstrate the use of ALTO in determining how to scale-out a VPN on demand. We also present a prototype of an ALTO-based dynamic VPN management, which is based on a carrier-grade network management system. Our results show that ALTO is a powerful topology abstraction approach that enables informed VPN scale-out decisions by applications.
{"title":"Dynamic VPN Optimization by ALTO Guidance","authors":"M. Scharf, V. Gurbani, T. Voith, M. Stein, W. D. Roome, Greg Soprovich, V. Hilt","doi":"10.1109/EWSDN.2013.9","DOIUrl":"https://doi.org/10.1109/EWSDN.2013.9","url":null,"abstract":"Virtual Private Networks (VPNs) are a key component of cloud computing systems, since they provide isolated connectivity between geographically separated users. The elasticity in cloud computing and new usage patterns such as cloud bursting require VPNs to be more dynamic than traditional solutions used by network service providers. Managing and optimizing the topology of VPNs requires insight into the underlying wide area network topology and benefits from new network interfaces currently discussed for Software Defined Networks (SDN). This paper presents the use of the Application-Layer Traffic Optimization (ALTO) protocol for VPN optimization. ALTO is a standardized solution for exposure of abstract topology information to a variety of applications, including cloud management systems. We demonstrate the use of ALTO in determining how to scale-out a VPN on demand. We also present a prototype of an ALTO-based dynamic VPN management, which is based on a carrier-grade network management system. Our results show that ALTO is a powerful topology abstraction approach that enables informed VPN scale-out decisions by applications.","PeriodicalId":256710,"journal":{"name":"2013 Second European Workshop on Software Defined Networks","volume":"472 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122235324","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}
Software-Defined Networking (SDN) promises the vision of more flexible and manageable networks, but requires certain level of programmability in the data plane. Such a flexible, programmable data plane implementation is OpenFlow (OF) which these days is seen as primary model of SDN data plane. In this paper we focus on the limitations of OF in packet switching performance. We share some measurement results we collected using an OF 1.3 prototype based on Intel's Data Plane Development Kit (DPDK) and we also describe some optimization ideas. While OF 1.0 can be implemented on high-speed Ethernet switch hardware it has certain disadvantages in the area of flexibility. On the other hand OF 1.3 offers good-enough flexibility, but the poor performance of OF 1.3 implementations seems to represent a roadblock to SDN adoption. In this paper we argue that contrast to the common view, the overhead of flexibility is relatively low. We also argue that the apparent difference between a programmable data plane and a state of the art layered data plane is not primarily due to flexibility itself, but because the lack of optimization in case of flexible implementations.
{"title":"Removing Roadblocks from SDN: OpenFlow Software Switch Performance on Intel DPDK","authors":"Gergely Pongrácz, László Molnár, Z. Kis","doi":"10.1109/EWSDN.2013.17","DOIUrl":"https://doi.org/10.1109/EWSDN.2013.17","url":null,"abstract":"Software-Defined Networking (SDN) promises the vision of more flexible and manageable networks, but requires certain level of programmability in the data plane. Such a flexible, programmable data plane implementation is OpenFlow (OF) which these days is seen as primary model of SDN data plane. In this paper we focus on the limitations of OF in packet switching performance. We share some measurement results we collected using an OF 1.3 prototype based on Intel's Data Plane Development Kit (DPDK) and we also describe some optimization ideas. While OF 1.0 can be implemented on high-speed Ethernet switch hardware it has certain disadvantages in the area of flexibility. On the other hand OF 1.3 offers good-enough flexibility, but the poor performance of OF 1.3 implementations seems to represent a roadblock to SDN adoption. In this paper we argue that contrast to the common view, the overhead of flexibility is relatively low. We also argue that the apparent difference between a programmable data plane and a state of the art layered data plane is not primarily due to flexibility itself, but because the lack of optimization in case of flexible implementations.","PeriodicalId":256710,"journal":{"name":"2013 Second European Workshop on Software Defined Networks","volume":"105 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134199631","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}
Jose Teixeira, G. Antichi, Alessio Del Chiaro, S. Giordano, Alexandre J. T. Santos
In the last years, the widespread of Cloud computing as the main paradigm to deliver a large plethora of virtualized services significantly increased the complexity of Datacenters management and raised new performance issues for the intra-Datacenter network. Providing heterogeneous services and satisfying users' experience is really challenging for Cloud service providers, since system (IT resources) and network administration functions are definitely separated. As the Software Defined Networking (SDN) approach seems to be a promising way to address innovation in Datacenters, the paper presents a new framework that allows to develop and test new OpenFlow -- based controllers for Cloud Datacenters. More specifically, our framework enhances both Mininet (a well -- known SDN emulator) and POX (a network controller written in python), with all the extensions necessary to experiment novel control and management strategies of IT and network resources.
{"title":"Datacenter in a Box: Test Your SDN Cloud-Datacenter Controller at Home","authors":"Jose Teixeira, G. Antichi, Alessio Del Chiaro, S. Giordano, Alexandre J. T. Santos","doi":"10.1109/EWSDN.2013.23","DOIUrl":"https://doi.org/10.1109/EWSDN.2013.23","url":null,"abstract":"In the last years, the widespread of Cloud computing as the main paradigm to deliver a large plethora of virtualized services significantly increased the complexity of Datacenters management and raised new performance issues for the intra-Datacenter network. Providing heterogeneous services and satisfying users' experience is really challenging for Cloud service providers, since system (IT resources) and network administration functions are definitely separated. As the Software Defined Networking (SDN) approach seems to be a promising way to address innovation in Datacenters, the paper presents a new framework that allows to develop and test new OpenFlow -- based controllers for Cloud Datacenters. More specifically, our framework enhances both Mininet (a well -- known SDN emulator) and POX (a network controller written in python), with all the extensions necessary to experiment novel control and management strategies of IT and network resources.","PeriodicalId":256710,"journal":{"name":"2013 Second European Workshop on Software Defined Networks","volume":"180 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122301336","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}
F. Botelho, Fernando M. V. Ramos, D. Kreutz, A. Bessani
Maintaining a strongly consistent network view in a Software Defined Network has been usually proclaimed as a synonym of low performance. We disagree. To support our view, in this paper we argue that with the use of modern distributed systems techniques it is possible to build a strongly consistent, fault-tolerant SDN control framework that achieves acceptable performance. The central element of our architecture is a highly-available, strongly consistent data store. We describe a prototype implementation of a distributed controller architecture integrating the Floodlight controller with a data store implemented using a state-of-the-art replication algorithm. We evaluate the feasibility of the proposed design by analyzing the workloads of real SDN applications (a learning switch, a load balancer and a device manager) and showing that the data store is capable of handling them with adequate performance.
{"title":"On the Feasibility of a Consistent and Fault-Tolerant Data Store for SDNs","authors":"F. Botelho, Fernando M. V. Ramos, D. Kreutz, A. Bessani","doi":"10.1109/EWSDN.2013.13","DOIUrl":"https://doi.org/10.1109/EWSDN.2013.13","url":null,"abstract":"Maintaining a strongly consistent network view in a Software Defined Network has been usually proclaimed as a synonym of low performance. We disagree. To support our view, in this paper we argue that with the use of modern distributed systems techniques it is possible to build a strongly consistent, fault-tolerant SDN control framework that achieves acceptable performance. The central element of our architecture is a highly-available, strongly consistent data store. We describe a prototype implementation of a distributed controller architecture integrating the Floodlight controller with a data store implemented using a state-of-the-art replication algorithm. We evaluate the feasibility of the proposed design by analyzing the workloads of real SDN applications (a learning switch, a load balancer and a device manager) and showing that the data store is capable of handling them with adequate performance.","PeriodicalId":256710,"journal":{"name":"2013 Second European Workshop on Software Defined Networks","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127474908","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}
Software Defined Networking (SDN) concepts are seen as suitable enablers for network virtualization, especially in the Data Center Network domain. However, also carrier network operators can benefit from network virtualization, since it allows new business models, promising economical benefits through sharing the cost of network infrastructure in e.g. multi-tenancy or service-isolation scenarios. Such use-cases pose additional requirements on virtualization schemes, including strict performance and information isolation, transparency of the virtualization system, high availability, as well as low CAPEX and OPEX demands. In order to fulfill these requirements, we previously proposed a flexible virtualization scheme for OpenFlow. In this paper we discuss the implementation of the proposed scheme and point out relevant lessons learned during the process, leading to architectural and technological updates. We then evaluate the system in terms of data path performance: the impact on forwarding latency is negligible, while the impact on network throughput is depending on the type of traffic and the choice of encapsulation technology. In summary, the overhead can be kept small and would not significantly affect a production network. Thus, we conclude that the minor performance degradations are outweighed by the benefits of the virtualization system.
{"title":"Implementation and Evaluation of a Carrier-Grade OpenFlow Virtualization Scheme","authors":"Pontus Sköldström, Wolfgang John","doi":"10.1109/EWSDN.2013.19","DOIUrl":"https://doi.org/10.1109/EWSDN.2013.19","url":null,"abstract":"Software Defined Networking (SDN) concepts are seen as suitable enablers for network virtualization, especially in the Data Center Network domain. However, also carrier network operators can benefit from network virtualization, since it allows new business models, promising economical benefits through sharing the cost of network infrastructure in e.g. multi-tenancy or service-isolation scenarios. Such use-cases pose additional requirements on virtualization schemes, including strict performance and information isolation, transparency of the virtualization system, high availability, as well as low CAPEX and OPEX demands. In order to fulfill these requirements, we previously proposed a flexible virtualization scheme for OpenFlow. In this paper we discuss the implementation of the proposed scheme and point out relevant lessons learned during the process, leading to architectural and technological updates. We then evaluate the system in terms of data path performance: the impact on forwarding latency is negligible, while the impact on network throughput is depending on the type of traffic and the choice of encapsulation technology. In summary, the overhead can be kept small and would not significantly affect a production network. Thus, we conclude that the minor performance degradations are outweighed by the benefits of the virtualization system.","PeriodicalId":256710,"journal":{"name":"2013 Second European Workshop on Software Defined Networks","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134539279","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：