Pub Date : 2020-11-10DOI: 10.1109/NFV-SDN50289.2020.9289867
M. Abranches, Eric Keller
While we cannot question the high performance capabilities of the kernel bypass approach in the network functions world, we recognize that the Linux kernel provides a rich ecosystem with an efficient resource management and an effective resource sharing ability that cannot be ignored. In this work we argue that by mixing kernel-bypass and in kernel processing can benefit applications and network function middleboxes. We leverage a high-performance user space TCP stack and recent additions to the Linux kernel to propose a hybrid approach (kernel-user space) to accelerate SDN/NFV deployments leveraging services of the reliable transport layer (i.e., stateful middleboxes, Layer 7 network functions and applications). Our results show that this approach enables highperformance, high CPU efficiency, and enhanced integration with the kernel ecosystem. We build our solution by extending mTCP which is the basis of some state-of-the-art L4-L7 NFV frameworks. By having more efficient CPU usage, NFV applications can have more CPU cycles available to run the network functions and applications logic. We show that for a CPU intense workload, mTCP/AF_XDP can have up to 64% more throughput than the previous implementation. We also show that by receiving cooperation from the kernel, mTCP/AF_XDP enables the creation of protection mechanisms for mTCP. We create a simulated DDoS attack and show that mTCP/AF_XDP can maintain up to 287% more throughput than the unprotected system during the attack.
{"title":"A Userspace Transport Stack Doesn't Have to Mean Losing Linux Processing","authors":"M. Abranches, Eric Keller","doi":"10.1109/NFV-SDN50289.2020.9289867","DOIUrl":"https://doi.org/10.1109/NFV-SDN50289.2020.9289867","url":null,"abstract":"While we cannot question the high performance capabilities of the kernel bypass approach in the network functions world, we recognize that the Linux kernel provides a rich ecosystem with an efficient resource management and an effective resource sharing ability that cannot be ignored. In this work we argue that by mixing kernel-bypass and in kernel processing can benefit applications and network function middleboxes. We leverage a high-performance user space TCP stack and recent additions to the Linux kernel to propose a hybrid approach (kernel-user space) to accelerate SDN/NFV deployments leveraging services of the reliable transport layer (i.e., stateful middleboxes, Layer 7 network functions and applications). Our results show that this approach enables highperformance, high CPU efficiency, and enhanced integration with the kernel ecosystem. We build our solution by extending mTCP which is the basis of some state-of-the-art L4-L7 NFV frameworks. By having more efficient CPU usage, NFV applications can have more CPU cycles available to run the network functions and applications logic. We show that for a CPU intense workload, mTCP/AF_XDP can have up to 64% more throughput than the previous implementation. We also show that by receiving cooperation from the kernel, mTCP/AF_XDP enables the creation of protection mechanisms for mTCP. We create a simulated DDoS attack and show that mTCP/AF_XDP can maintain up to 287% more throughput than the unprotected system during the attack.","PeriodicalId":283280,"journal":{"name":"2020 IEEE Conference on Network Function Virtualization and Software Defined Networks (NFV-SDN)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126015625","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 : 2020-11-10DOI: 10.1109/nfv-sdn50289.2020.9289897
Steven L. Wolf, Conor Walsh, M. C. Cavusoglu, Tianming Liu, Ferdinando Rodriguez Y Baena
{"title":"Keynote Talks","authors":"Steven L. Wolf, Conor Walsh, M. C. Cavusoglu, Tianming Liu, Ferdinando Rodriguez Y Baena","doi":"10.1109/nfv-sdn50289.2020.9289897","DOIUrl":"https://doi.org/10.1109/nfv-sdn50289.2020.9289897","url":null,"abstract":"","PeriodicalId":283280,"journal":{"name":"2020 IEEE Conference on Network Function Virtualization and Software Defined Networks (NFV-SDN)","volume":"148 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123301443","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 : 2020-11-10DOI: 10.1109/NFV-SDN50289.2020.9289861
R. Vilalta, P. Alemany, R. Sedar, Charalampos Kalalas, R. Casellas, R. Martínez, F. V. Gallego, Jordi Ortiz Murillo, A. Gómez-Skarmeta, J. Alonso-Zarate, R. Muñoz
This demo presents the instantiation of a high level architecture proposed by INSPIRE-5Gplus to manage secured End-to-End (E2E) Network Slices. Network Slices are associated with Security Service Level Agreement (SSLA) to enhance the security on the virtual deployed resources, thus securing the created E2E Secure Network Slices. The proposed secured network slicing architecture is validated against a vehicular scenario, based on Anticipated Cooperative Collision Avoidance use case. In this scenario, we propose to detect false vehicular messages through a novel SSLA based on an Intrusion Detection System. The demo provides measurement of significant metrics such as mean time to detect, mean time to contain, and transaction speed.
本演示演示了INSPIRE-5Gplus提出的用于管理安全的端到端(E2E)网络切片的高级架构的实例化。通过将网络切片与SSLA (Security Service Level Agreement)协议关联，增强虚拟部署资源的安全性，从而保证创建的端到端安全网络切片的安全性。基于预期协同避碰用例，针对车辆场景验证了所提出的安全网络切片架构。在这种情况下，我们建议通过基于入侵检测系统的新型SSLA来检测虚假车辆信息。该演示提供了重要指标的度量，例如平均检测时间、平均包含时间和事务速度。
{"title":"Applying Security Service Level Agreements in V2X Network Slices","authors":"R. Vilalta, P. Alemany, R. Sedar, Charalampos Kalalas, R. Casellas, R. Martínez, F. V. Gallego, Jordi Ortiz Murillo, A. Gómez-Skarmeta, J. Alonso-Zarate, R. Muñoz","doi":"10.1109/NFV-SDN50289.2020.9289861","DOIUrl":"https://doi.org/10.1109/NFV-SDN50289.2020.9289861","url":null,"abstract":"This demo presents the instantiation of a high level architecture proposed by INSPIRE-5Gplus to manage secured End-to-End (E2E) Network Slices. Network Slices are associated with Security Service Level Agreement (SSLA) to enhance the security on the virtual deployed resources, thus securing the created E2E Secure Network Slices. The proposed secured network slicing architecture is validated against a vehicular scenario, based on Anticipated Cooperative Collision Avoidance use case. In this scenario, we propose to detect false vehicular messages through a novel SSLA based on an Intrusion Detection System. The demo provides measurement of significant metrics such as mean time to detect, mean time to contain, and transaction speed.","PeriodicalId":283280,"journal":{"name":"2020 IEEE Conference on Network Function Virtualization and Software Defined Networks (NFV-SDN)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124149877","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 : 2020-11-10DOI: 10.1109/NFV-SDN50289.2020.9289855
Alessandra Fais, S. Giordano, G. Procissi
The current trend characterizing new generation networks is to accommodate a variety of services on the same shared infrastructure. In this context, two main aspects are vital for network operators. On one hand, there is the need for proper mechanisms to rapidly and easily (re-)configure the network, to adapt to its changing conditions over time. On the other hand, applications for continuous network monitoring become essential to detect security problems or performance degradation. This is fundamental to guarantee the Quality of Service (QoS) requirements for every running service. This paper aims at proposing a new framework for the implementation of fast and scalable applications for real-time continuous network monitoring and data analysis. High-level abstractions will be provided to the network programmer, and the Data Stream Processing computational model will be exploited to improve performance. The framework architecture is described, along with the implementation design for this kind of applications. The main challenges are presented, with proposed solutions to tackle them. Finally, the current status of the work is discussed, along with its future developments.
{"title":"On the Design of Fast and Scalable Network Applications Through Data Stream Processing","authors":"Alessandra Fais, S. Giordano, G. Procissi","doi":"10.1109/NFV-SDN50289.2020.9289855","DOIUrl":"https://doi.org/10.1109/NFV-SDN50289.2020.9289855","url":null,"abstract":"The current trend characterizing new generation networks is to accommodate a variety of services on the same shared infrastructure. In this context, two main aspects are vital for network operators. On one hand, there is the need for proper mechanisms to rapidly and easily (re-)configure the network, to adapt to its changing conditions over time. On the other hand, applications for continuous network monitoring become essential to detect security problems or performance degradation. This is fundamental to guarantee the Quality of Service (QoS) requirements for every running service. This paper aims at proposing a new framework for the implementation of fast and scalable applications for real-time continuous network monitoring and data analysis. High-level abstractions will be provided to the network programmer, and the Data Stream Processing computational model will be exploited to improve performance. The framework architecture is described, along with the implementation design for this kind of applications. The main challenges are presented, with proposed solutions to tackle them. Finally, the current status of the work is discussed, along with its future developments.","PeriodicalId":283280,"journal":{"name":"2020 IEEE Conference on Network Function Virtualization and Software Defined Networks (NFV-SDN)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116217074","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 : 2020-11-10DOI: 10.1109/NFV-SDN50289.2020.9289860
Christian Wernecke, Helge Parzyjegla, Gero Mühl
In content-based publish/subscribe systems, publishers produce notifications and subscribers register subscriptions to receive certain notifications based on their content. It is the task of the middleware to deliver the published notifications to subscribers having a matching subscription. Content-based publish/subscribe is challenging to implement efficiently, because the resulting delivery tree can be different for each notification, set of active subscriptions, and location of the subscribers. Therefore, content-based routing of notification has often been implemented in the application layer resulting in high notification delays and a waste of network bandwidth. In this work, we report on the research and actual development of a novel middleware, which uses the P4 SDN programming language to perform the content-based forwarding of notifications solely in the network layer. We present a hybrid approach that combines (i) encoding of routing information into the notification to be distributed and (ii) installing forwarding rules into the switches. In particular, we combine different types of routing information that either represent individual network links or virtual links referencing pre-installed distribution trees. We derive the latter from topological information about the network and from publish/subscribe relationships as well as forwarding statistics.
{"title":"Implementing Content-based Publish/Subscribe on the Network Layer with P4","authors":"Christian Wernecke, Helge Parzyjegla, Gero Mühl","doi":"10.1109/NFV-SDN50289.2020.9289860","DOIUrl":"https://doi.org/10.1109/NFV-SDN50289.2020.9289860","url":null,"abstract":"In content-based publish/subscribe systems, publishers produce notifications and subscribers register subscriptions to receive certain notifications based on their content. It is the task of the middleware to deliver the published notifications to subscribers having a matching subscription. Content-based publish/subscribe is challenging to implement efficiently, because the resulting delivery tree can be different for each notification, set of active subscriptions, and location of the subscribers. Therefore, content-based routing of notification has often been implemented in the application layer resulting in high notification delays and a waste of network bandwidth. In this work, we report on the research and actual development of a novel middleware, which uses the P4 SDN programming language to perform the content-based forwarding of notifications solely in the network layer. We present a hybrid approach that combines (i) encoding of routing information into the notification to be distributed and (ii) installing forwarding rules into the switches. In particular, we combine different types of routing information that either represent individual network links or virtual links referencing pre-installed distribution trees. We derive the latter from topological information about the network and from publish/subscribe relationships as well as forwarding statistics.","PeriodicalId":283280,"journal":{"name":"2020 IEEE Conference on Network Function Virtualization and Software Defined Networks (NFV-SDN)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121510548","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 : 2020-11-10DOI: 10.1109/NFV-SDN50289.2020.9289873
Daniele Moro, D. Sanvito, A. Capone
Extended Finite State Machines (EFSMs) are a very promising abstraction to describe stateful networking applications. However, the community currently lacks a fast prototyping environment to experiment with EFSM-based data plane applications. In this demonstration, we present FlowBlaze.p4, an EFSM executor based on P4. The executor can be run on Mininet and BMv2, helping to prototype new use cases with a small turnaround time. We provide a set of tools, including a GUI, to simplify the development by interactively drawing the state machines. We demonstrate FlowBlaze.p4 with two example scenarios, showing how developers can exploit the proposed library to implement their own use cases.
{"title":"Demonstrating FlowBlaze.p4: fast prototyping for EFSM-based data plane applications","authors":"Daniele Moro, D. Sanvito, A. Capone","doi":"10.1109/NFV-SDN50289.2020.9289873","DOIUrl":"https://doi.org/10.1109/NFV-SDN50289.2020.9289873","url":null,"abstract":"Extended Finite State Machines (EFSMs) are a very promising abstraction to describe stateful networking applications. However, the community currently lacks a fast prototyping environment to experiment with EFSM-based data plane applications. In this demonstration, we present FlowBlaze.p4, an EFSM executor based on P4. The executor can be run on Mininet and BMv2, helping to prototype new use cases with a small turnaround time. We provide a set of tools, including a GUI, to simplify the development by interactively drawing the state machines. We demonstrate FlowBlaze.p4 with two example scenarios, showing how developers can exploit the proposed library to implement their own use cases.","PeriodicalId":283280,"journal":{"name":"2020 IEEE Conference on Network Function Virtualization and Software Defined Networks (NFV-SDN)","volume":"68 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128637459","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 : 2020-11-10DOI: 10.1109/NFV-SDN50289.2020.9289877
Sidharth Sharma, A. Gumaste
The next-generation software-defined networks (SDNs) are designed to be extremely flexible and cost-effective. However, complying with service-level agreements (SLAs) is often challenging with the softwarization of networks. To this end, we propose different methods and techniques in this paper to solve some crucial problems related to flow provisioning with SLA guarantees. The paper is divided into two parts: The first part focuses on the SLA-driven design of SDNs. First, we propose Bitstream, a programmable SDN protocol that can provide carrier-class features. We demonstrate Bitstream on the test-bed of custom-built FPGA-based Terabit white-boxes. We show the effectiveness of Bitstream in providing low latency and 50ms restoration in case of faults. Next, we propose techniques to make Bitstream more application-centric by including delay requirements of service requests. In the second part, we shift our focus to Network Function Virtualization (NFV). One of the biggest hurdles of adopting NFV is to provide a similar performance what a service provider is accustomed to with the traditional network hardware. First, we design an end-to-end service provider network with NFV data-centers. To provide highly-available services, we create an ILP framework intending to maximize the service provider's profit. We then propose a robust optimization extension for handling traffic uncertainty. Next, we propose a service function chain (SFC) realization theory that compares the requirements of an SFC request to the available resources in NFV infrastructure while providing availability and delay guarantees for the dynamic arrival of requests. Finally, we present an NFV use-case in the virtualized slicing of 5G networks. We propose a dynamic network slicing algorithm that creates/deletes/modifies slices while optimizing underlying virtualized resources. To handle uncertainty in the traffic requests, a stochastic optimization extension is also proposed.
{"title":"SLA-Aware Flow Provisioning in Next-Generation Software-Defined Networks","authors":"Sidharth Sharma, A. Gumaste","doi":"10.1109/NFV-SDN50289.2020.9289877","DOIUrl":"https://doi.org/10.1109/NFV-SDN50289.2020.9289877","url":null,"abstract":"The next-generation software-defined networks (SDNs) are designed to be extremely flexible and cost-effective. However, complying with service-level agreements (SLAs) is often challenging with the softwarization of networks. To this end, we propose different methods and techniques in this paper to solve some crucial problems related to flow provisioning with SLA guarantees. The paper is divided into two parts: The first part focuses on the SLA-driven design of SDNs. First, we propose Bitstream, a programmable SDN protocol that can provide carrier-class features. We demonstrate Bitstream on the test-bed of custom-built FPGA-based Terabit white-boxes. We show the effectiveness of Bitstream in providing low latency and 50ms restoration in case of faults. Next, we propose techniques to make Bitstream more application-centric by including delay requirements of service requests. In the second part, we shift our focus to Network Function Virtualization (NFV). One of the biggest hurdles of adopting NFV is to provide a similar performance what a service provider is accustomed to with the traditional network hardware. First, we design an end-to-end service provider network with NFV data-centers. To provide highly-available services, we create an ILP framework intending to maximize the service provider's profit. We then propose a robust optimization extension for handling traffic uncertainty. Next, we propose a service function chain (SFC) realization theory that compares the requirements of an SFC request to the available resources in NFV infrastructure while providing availability and delay guarantees for the dynamic arrival of requests. Finally, we present an NFV use-case in the virtualized slicing of 5G networks. We propose a dynamic network slicing algorithm that creates/deletes/modifies slices while optimizing underlying virtualized resources. To handle uncertainty in the traffic requests, a stochastic optimization extension is also proposed.","PeriodicalId":283280,"journal":{"name":"2020 IEEE Conference on Network Function Virtualization and Software Defined Networks (NFV-SDN)","volume":"53 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116777429","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 : 2020-11-10DOI: 10.1109/nfv-sdn50289.2020.9289878
Deval Bhamare
dedicated PhD which is taking place under the umbrella of the IEEE NFV-SDN conference which is in its sixth year. The NFV-SDN Doctoral Symposium is committed to fostering collaboration amongst PhD students and experts from all communities researching and working in the areas of Network Function Virtualization (NFV) and Software Defined Networks (SDN). It offers a unique opportunity for PhD students to present their latest research results, discuss new research ideas, and to gather valuable expert feedback on their work from experienced researchers from both industry and academia. Moreover, it is a place for mentoring and to get in touch with student peers working in the same field. The Doctoral Symposium invited submissions from PhD students at the beginning of their work when they have a well-defined problem statement and are developing their ideas on how to approach the challenges. Likewise, we also invited contributions from PhD students at a more advanced stage to present their research results and share insights coming from their experience which might help others to progress their work. We received a total of 14 submissions of which one early-stage and five later-stage contributions were selected for presentation. The selected papers are organized in two sessions addressing current hot topics ranging from performance and security aspects of virtual network functions with network slicing and slice management, to the seamless integration of network services into software-defined networks with and without programmable data planes. We would like to all the authors for their contributions; all the people involved in the Doctoral Symposium Program Committee for their dedication; all the members of the IEEE NFV-SDN Organizing Committee for their encouragement and support; and the Local Organizing Team of Universidad III de Madrid for their tremendous efforts to transform the symposium, originally planned as an on-location workshop, into a sophisticated virtual event because of the ongoing pandemic.
{"title":"NFV-SDN Doctoral Symposium","authors":"Deval Bhamare","doi":"10.1109/nfv-sdn50289.2020.9289878","DOIUrl":"https://doi.org/10.1109/nfv-sdn50289.2020.9289878","url":null,"abstract":"dedicated PhD which is taking place under the umbrella of the IEEE NFV-SDN conference which is in its sixth year. The NFV-SDN Doctoral Symposium is committed to fostering collaboration amongst PhD students and experts from all communities researching and working in the areas of Network Function Virtualization (NFV) and Software Defined Networks (SDN). It offers a unique opportunity for PhD students to present their latest research results, discuss new research ideas, and to gather valuable expert feedback on their work from experienced researchers from both industry and academia. Moreover, it is a place for mentoring and to get in touch with student peers working in the same field. The Doctoral Symposium invited submissions from PhD students at the beginning of their work when they have a well-defined problem statement and are developing their ideas on how to approach the challenges. Likewise, we also invited contributions from PhD students at a more advanced stage to present their research results and share insights coming from their experience which might help others to progress their work. We received a total of 14 submissions of which one early-stage and five later-stage contributions were selected for presentation. The selected papers are organized in two sessions addressing current hot topics ranging from performance and security aspects of virtual network functions with network slicing and slice management, to the seamless integration of network services into software-defined networks with and without programmable data planes. We would like to all the authors for their contributions; all the people involved in the Doctoral Symposium Program Committee for their dedication; all the members of the IEEE NFV-SDN Organizing Committee for their encouragement and support; and the Local Organizing Team of Universidad III de Madrid for their tremendous efforts to transform the symposium, originally planned as an on-location workshop, into a sophisticated virtual event because of the ongoing pandemic.","PeriodicalId":283280,"journal":{"name":"2020 IEEE Conference on Network Function Virtualization and Software Defined Networks (NFV-SDN)","volume":"1948 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128817206","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 : 2020-11-10DOI: 10.1109/NFV-SDN50289.2020.9289902
Min Xie, P. Gomes, J. Harmatos, Jose A. Ordonez-Lucena
In 5G networks of high complexity, autonomous management of the end-to-end (E2E) services is desired. Closed-loop (CL) control is a future-proof means to enable this autonomy. In the multi-provider, multi-vendor, and multi-tenant environment, the autonomous service management demands collaborations of all providers and vendors. In this paper, the collaboration problem is formulated as cross-CL collaboration involving multiple CLs participating in the E2E service management. We present a cross-CL collaboration framework with key elements, e.g., APIs, governance models, intent and policies management. Then a representative use case of service assurance with edge clouds is described to exemplify how the framework is applied to implement cross-CL collaborations and answer key questions: WHEN collaboration is needed; and HOW collaboration is enabled.
{"title":"Collaborated Closed Loops for Autonomous End-to-End Service Management in 5G","authors":"Min Xie, P. Gomes, J. Harmatos, Jose A. Ordonez-Lucena","doi":"10.1109/NFV-SDN50289.2020.9289902","DOIUrl":"https://doi.org/10.1109/NFV-SDN50289.2020.9289902","url":null,"abstract":"In 5G networks of high complexity, autonomous management of the end-to-end (E2E) services is desired. Closed-loop (CL) control is a future-proof means to enable this autonomy. In the multi-provider, multi-vendor, and multi-tenant environment, the autonomous service management demands collaborations of all providers and vendors. In this paper, the collaboration problem is formulated as cross-CL collaboration involving multiple CLs participating in the E2E service management. We present a cross-CL collaboration framework with key elements, e.g., APIs, governance models, intent and policies management. Then a representative use case of service assurance with edge clouds is described to exemplify how the framework is applied to implement cross-CL collaborations and answer key questions: WHEN collaboration is needed; and HOW collaboration is enabled.","PeriodicalId":283280,"journal":{"name":"2020 IEEE Conference on Network Function Virtualization and Software Defined Networks (NFV-SDN)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127527299","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 : 2020-11-10DOI: 10.1109/NFV-SDN50289.2020.9289901
Anees Ullah, S. Pontarelli, P. Reviriego
Field Programmable Gate Arrays (FPGAs) have evolved into large system on chip devices that have high speed network interfaces, processors and vast amounts of logic and memory blocks. This coupled with their inherent reconfigurability, makes them an attractive platform to implement Software Defined Networking (SDN) systems. However, there is one piece missing that is used in most networking devices: Ternary Content Addressable Memories (TCAMs). TCAMs are widely used to implement packet classification. TCAMs can match an incoming key against a set of stored rules that include wildcard bits in one memory access. TCAMs are usually designed as specific hardware blocks and are included in almost all switching ASICs. However, they are not implemented in FPGA devices. This is due to the cost in terms of area/power of TCAMs hardware blocks and their lack of flexibility. To provide packet classification in FPGA based systems, a large number of schemes to emulate TCAMs on FPGAs have been proposed in recent years. In many cases, they are based on the use of the memory blocks available on the FPGA. In this paper, it is shown that FPGA emulated TCAMs are actually more powerful than traditional TCAMs and can support more generic rules. This makes them more efficient in some applications as the number of rules needed to implement a system can be significantly reduced. This is illustrated with two case studies: supporting port lists and ranges in access control lists and blacklisting of IP addresses. The results show the benefits of using more generic rules in the FPGA emulated TCAMs.
{"title":"FlexTCAM: Beyond Memory Based TCAM Emulation on FPGAs","authors":"Anees Ullah, S. Pontarelli, P. Reviriego","doi":"10.1109/NFV-SDN50289.2020.9289901","DOIUrl":"https://doi.org/10.1109/NFV-SDN50289.2020.9289901","url":null,"abstract":"Field Programmable Gate Arrays (FPGAs) have evolved into large system on chip devices that have high speed network interfaces, processors and vast amounts of logic and memory blocks. This coupled with their inherent reconfigurability, makes them an attractive platform to implement Software Defined Networking (SDN) systems. However, there is one piece missing that is used in most networking devices: Ternary Content Addressable Memories (TCAMs). TCAMs are widely used to implement packet classification. TCAMs can match an incoming key against a set of stored rules that include wildcard bits in one memory access. TCAMs are usually designed as specific hardware blocks and are included in almost all switching ASICs. However, they are not implemented in FPGA devices. This is due to the cost in terms of area/power of TCAMs hardware blocks and their lack of flexibility. To provide packet classification in FPGA based systems, a large number of schemes to emulate TCAMs on FPGAs have been proposed in recent years. In many cases, they are based on the use of the memory blocks available on the FPGA. In this paper, it is shown that FPGA emulated TCAMs are actually more powerful than traditional TCAMs and can support more generic rules. This makes them more efficient in some applications as the number of rules needed to implement a system can be significantly reduced. This is illustrated with two case studies: supporting port lists and ranges in access control lists and blacklisting of IP addresses. The results show the benefits of using more generic rules in the FPGA emulated TCAMs.","PeriodicalId":283280,"journal":{"name":"2020 IEEE Conference on Network Function Virtualization and Software Defined Networks (NFV-SDN)","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131425331","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：