Pub Date : 2020-06-01DOI: 10.1109/NetSoft48620.2020.9165380
Abdunasser Alowa, T. Fevens
Software Defined Networking (SDN) is a recently proposed networking pattern that facilitates a centralized system of computer networks where a controller maintains the management of a global view of the network. One of the characteristics of SDNs is that by decoupling the control and data plane from each other, the controllability and manageability of a network is improved. In this arrangement, the connections between the controller (control plane) and the switches (data plane) are established by either an in-band or an out-of-band control mechanism. Despite all the advantages of SDNs, new challenges arise regarding the connection availability between the data and control planes. A disconnection between the two planes could result in performance degradation. To achieve reliable control traffic between data and control planes, in this work, we design and implement an in-band control protection approach that finds a set of ideal paths for control channel, where as much control traffic as possible can be protected by the proposed protection mechanism. This design enables switches to locally react to failures without involving the controller. Through simulation experiments, we show that our proposed approach significantly improves the control reliability of an in-band control network.
{"title":"A Dynamic Recovery Module for In-band Control Channel Failure In Software Defined Networking","authors":"Abdunasser Alowa, T. Fevens","doi":"10.1109/NetSoft48620.2020.9165380","DOIUrl":"https://doi.org/10.1109/NetSoft48620.2020.9165380","url":null,"abstract":"Software Defined Networking (SDN) is a recently proposed networking pattern that facilitates a centralized system of computer networks where a controller maintains the management of a global view of the network. One of the characteristics of SDNs is that by decoupling the control and data plane from each other, the controllability and manageability of a network is improved. In this arrangement, the connections between the controller (control plane) and the switches (data plane) are established by either an in-band or an out-of-band control mechanism. Despite all the advantages of SDNs, new challenges arise regarding the connection availability between the data and control planes. A disconnection between the two planes could result in performance degradation. To achieve reliable control traffic between data and control planes, in this work, we design and implement an in-band control protection approach that finds a set of ideal paths for control channel, where as much control traffic as possible can be protected by the proposed protection mechanism. This design enables switches to locally react to failures without involving the controller. Through simulation experiments, we show that our proposed approach significantly improves the control reliability of an in-band control network.","PeriodicalId":239961,"journal":{"name":"2020 6th IEEE Conference on Network Softwarization (NetSoft)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128767960","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 : 2020-06-01DOI: 10.1109/NetSoft48620.2020.9165521
A. Al-Jawad, P. Shah, O. Gemikonakli, I. Comsa, R. Trestian
Most of the existing performance evaluation studies of various routing algorithms are done under limited experimental setups leading to an incomplete picture of the routing algorithm performance under dynamic network conditions. This paper presents a study that compares state-of-the-art routing algorithms over realistic multimedia-based Software Defined Networks (SDNs) with dynamic network conditions and various topology. Routing algorithms remain a key element of the networking landscape as they determine the path the data packets follow. The next-generation networking paradigm offers wide advantages over traditional networks through simplifying the management layer, especially with the adoption of SDN. However, Quality of Service (QoS) provisioning still remains a challenge that needs to be investigated especially for multimedia-based SDNs. This study investigates the impact of state-of-the-art centralized routing algorithms (e.g. MHA, WSP, SWP, MIRA) on multimedia QoS traffic under a realistic environment in terms of PSNR, Throughput, Packet Loss, Delay and QoS rejection.
{"title":"Performance Evaluation of Routing Strategies over Multimedia-based SDNs under Realistic Environments","authors":"A. Al-Jawad, P. Shah, O. Gemikonakli, I. Comsa, R. Trestian","doi":"10.1109/NetSoft48620.2020.9165521","DOIUrl":"https://doi.org/10.1109/NetSoft48620.2020.9165521","url":null,"abstract":"Most of the existing performance evaluation studies of various routing algorithms are done under limited experimental setups leading to an incomplete picture of the routing algorithm performance under dynamic network conditions. This paper presents a study that compares state-of-the-art routing algorithms over realistic multimedia-based Software Defined Networks (SDNs) with dynamic network conditions and various topology. Routing algorithms remain a key element of the networking landscape as they determine the path the data packets follow. The next-generation networking paradigm offers wide advantages over traditional networks through simplifying the management layer, especially with the adoption of SDN. However, Quality of Service (QoS) provisioning still remains a challenge that needs to be investigated especially for multimedia-based SDNs. This study investigates the impact of state-of-the-art centralized routing algorithms (e.g. MHA, WSP, SWP, MIRA) on multimedia QoS traffic under a realistic environment in terms of PSNR, Throughput, Packet Loss, Delay and QoS rejection.","PeriodicalId":239961,"journal":{"name":"2020 6th IEEE Conference on Network Softwarization (NetSoft)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126661844","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 : 2020-06-01DOI: 10.1109/NetSoft48620.2020.9165342
S. Clayman, F. Tusa, A. Galis, L. Contreras
This paper explores and makes a case for allocating a Wide-area Infrastructure Manager (WIM) on-demand to support softwarized network slicing, as part of the full NFVI virtualised infrastructure foundation, to ensure that the connectivity attributes prescribed to network slices can be managed with flexibility and adaptability in a full end-to-end slice. We show how creating a WIM on-demand and dynamically allocating a new WIM for each network slice, rather than having one for the whole network, can be beneficial for various slicing scenarios, in a similar way that a Virtual Infrastructure Manager (VIM) on-demand has been utilized. The paper considers some of the components, abstractions, and mechanisms of WIM on-demand.
{"title":"WIM on-demand – A modular approach for managing network slices","authors":"S. Clayman, F. Tusa, A. Galis, L. Contreras","doi":"10.1109/NetSoft48620.2020.9165342","DOIUrl":"https://doi.org/10.1109/NetSoft48620.2020.9165342","url":null,"abstract":"This paper explores and makes a case for allocating a Wide-area Infrastructure Manager (WIM) on-demand to support softwarized network slicing, as part of the full NFVI virtualised infrastructure foundation, to ensure that the connectivity attributes prescribed to network slices can be managed with flexibility and adaptability in a full end-to-end slice. We show how creating a WIM on-demand and dynamically allocating a new WIM for each network slice, rather than having one for the whole network, can be beneficial for various slicing scenarios, in a similar way that a Virtual Infrastructure Manager (VIM) on-demand has been utilized. The paper considers some of the components, abstractions, and mechanisms of WIM on-demand.","PeriodicalId":239961,"journal":{"name":"2020 6th IEEE Conference on Network Softwarization (NetSoft)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121817861","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 : 2020-06-01DOI: 10.1109/NetSoft48620.2020.9165471
Bin Li, Yipeng Wang, Ren Wang, Charlie Tai, R. Iyer, Zhu Zhou, Andrew J. Herdrich, Tong Zhang, Ameer Haj-Ali, I. Stoica, K. Asanović
Network function virtualization (NFV) technology attracts tremendous interests from telecommunication industry and data center operators, as it allows service providers to assign resource for Virtual Network Functions (VNFs) on demand, achieving better flexibility, programmability, and scalability. To improve server utilization, one popular practice is to deploy best effort (BE) workloads along with high priority (HP) VNFs when high priority VNF's resource usage is detected to be low. The key challenge of this deployment scheme is to dynamically balance the Service level objective (SLO) and the total cost of ownership (TCO) to optimize the data center efficiency under inherently fluctuating workloads. With the recent advancement in deep reinforcement learning, we conjecture that it has the potential to solve this challenge by adaptively adjusting resource allocation to reach the improved performance and higher server utilization. In this paper, we present a closed-loop automation system RLDRM11RLDRM: Reinforcement Learning Dynamic Resource Management to dynamically adjust Last Level Cache allocation between HP VNFs and BE workloads using deep reinforcement learning. The results demonstrate improved server utilization while maintaining required SLO for the HP VNFs.
{"title":"RLDRM: Closed Loop Dynamic Cache Allocation with Deep Reinforcement Learning for Network Function Virtualization","authors":"Bin Li, Yipeng Wang, Ren Wang, Charlie Tai, R. Iyer, Zhu Zhou, Andrew J. Herdrich, Tong Zhang, Ameer Haj-Ali, I. Stoica, K. Asanović","doi":"10.1109/NetSoft48620.2020.9165471","DOIUrl":"https://doi.org/10.1109/NetSoft48620.2020.9165471","url":null,"abstract":"Network function virtualization (NFV) technology attracts tremendous interests from telecommunication industry and data center operators, as it allows service providers to assign resource for Virtual Network Functions (VNFs) on demand, achieving better flexibility, programmability, and scalability. To improve server utilization, one popular practice is to deploy best effort (BE) workloads along with high priority (HP) VNFs when high priority VNF's resource usage is detected to be low. The key challenge of this deployment scheme is to dynamically balance the Service level objective (SLO) and the total cost of ownership (TCO) to optimize the data center efficiency under inherently fluctuating workloads. With the recent advancement in deep reinforcement learning, we conjecture that it has the potential to solve this challenge by adaptively adjusting resource allocation to reach the improved performance and higher server utilization. In this paper, we present a closed-loop automation system RLDRM11RLDRM: Reinforcement Learning Dynamic Resource Management to dynamically adjust Last Level Cache allocation between HP VNFs and BE workloads using deep reinforcement learning. The results demonstrate improved server utilization while maintaining required SLO for the HP VNFs.","PeriodicalId":239961,"journal":{"name":"2020 6th IEEE Conference on Network Softwarization (NetSoft)","volume":"475 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121398465","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}
The decomposition of a 5G radio access network (RAN) accelerates inter-operator infrastructure sharing of the corresponding transport network as well as the centralized deployment of a centralized unit (CU) and distributed unit (DU). We propose an architecture that enables inter-operator infrastructure sharing of 5G xHaul as slices and on-demand creation of xHaul network slices dynamically over multiple network operators that are interconnected and orchestrated with the standardized interface of the Open Networking Foundation (ONF) Transport API (TAPI) and corresponding standard from Metro Ethernet Forum (MEF). Through a proof of concept, we validated the implementation of the architecture in three live network domains for 5G xHaul.
{"title":"5G xHaul Sharing as Slice Implementation with inter- and intra-operator orchestration","authors":"Hiroki Baba, Takayuki Nakamura, A. Fukuda, Hiroshige Tanaka, Taiki Yamazaki, Noriyoshi Yamazaki, Noritaka Abe","doi":"10.1109/NetSoft48620.2020.9165514","DOIUrl":"https://doi.org/10.1109/NetSoft48620.2020.9165514","url":null,"abstract":"The decomposition of a 5G radio access network (RAN) accelerates inter-operator infrastructure sharing of the corresponding transport network as well as the centralized deployment of a centralized unit (CU) and distributed unit (DU). We propose an architecture that enables inter-operator infrastructure sharing of 5G xHaul as slices and on-demand creation of xHaul network slices dynamically over multiple network operators that are interconnected and orchestrated with the standardized interface of the Open Networking Foundation (ONF) Transport API (TAPI) and corresponding standard from Metro Ethernet Forum (MEF). Through a proof of concept, we validated the implementation of the architecture in three live network domains for 5G xHaul.","PeriodicalId":239961,"journal":{"name":"2020 6th IEEE Conference on Network Softwarization (NetSoft)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131628375","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 : 2020-06-01DOI: 10.1109/NetSoft48620.2020.9165458
S. Kukliński, Lechosław Tomaszewski, Pawel Korzec, Robert Kołakowski
The paper consists of a description of 5G-UASP, a multi-provider platform for drone services. It takes into account the involvement of air traffic management authority in all flight phases. We describe the core components of 5G-UASP and outline the procedures related to flight preparation, execution and termination phases. Instantiation of the concept is presented for 5G network. As it is shown, the usage of network slicing and MEC enables an immersive implementation of the architecture. Two options of instantiation of the reference architecture are included.
{"title":"5G-UASP: 5G-based multi-provider UAV platform architecture","authors":"S. Kukliński, Lechosław Tomaszewski, Pawel Korzec, Robert Kołakowski","doi":"10.1109/NetSoft48620.2020.9165458","DOIUrl":"https://doi.org/10.1109/NetSoft48620.2020.9165458","url":null,"abstract":"The paper consists of a description of 5G-UASP, a multi-provider platform for drone services. It takes into account the involvement of air traffic management authority in all flight phases. We describe the core components of 5G-UASP and outline the procedures related to flight preparation, execution and termination phases. Instantiation of the concept is presented for 5G network. As it is shown, the usage of network slicing and MEC enables an immersive implementation of the architecture. Two options of instantiation of the reference architecture are included.","PeriodicalId":239961,"journal":{"name":"2020 6th IEEE Conference on Network Softwarization (NetSoft)","volume":"186 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131886023","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 : 2020-06-01DOI: 10.1109/netsoft48620.2020.9165355
F. Turck
NetSoft 2020 is addressing network softwarization technologies, Cloud, fog and edge infrastructures for 5G, network programmability and software-defined networking, abstraction and virtualization of resources, services and functions, network slicing and slice orchestration, AI-based software-defined networks, and Open Source methodologies for network softwarization, among others. The topics of the conference are considered under the theme: “Bridging the gap between AI and Network Softwarization”.
{"title":"Message from the NetSoft 2020 Chairs","authors":"F. Turck","doi":"10.1109/netsoft48620.2020.9165355","DOIUrl":"https://doi.org/10.1109/netsoft48620.2020.9165355","url":null,"abstract":"NetSoft 2020 is addressing network softwarization technologies, Cloud, fog and edge infrastructures for 5G, network programmability and software-defined networking, abstraction and virtualization of resources, services and functions, network slicing and slice orchestration, AI-based software-defined networks, and Open Source methodologies for network softwarization, among others. The topics of the conference are considered under the theme: “Bridging the gap between AI and Network Softwarization”.","PeriodicalId":239961,"journal":{"name":"2020 6th IEEE Conference on Network Softwarization (NetSoft)","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134439680","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 : 2020-06-01DOI: 10.1109/NetSoft48620.2020.9165443
D. A. L. Perez, Christian Esteve Rothenberg, Mateus A. S. Santos, P. Gomes
Scenarios for distributed cloud with multiple edge clouds and centralized data centers are being investigated as the computing and networking underpinnings of next-generation network services such as augmented reality, self-driving vehicles, drones, and more. In such distributed environments, service providers will typically face tens, hundreds, or thousands of compute location candidates (edge, regional, and central) where network service components can be placed. To take optimized placement decisions of network services and execute the management workflows, orchestration systems require up-to-date and accurate resource availability representation, in the form of a network inventory that can be immense in distributed cloud scenarios. As a result, the service management and placement problems may become not tractable. In this work, we propose the Abstracted Network Inventory (ANI) component to generate service-optimized network views over the same network inventory. ANI implements a novel abstraction method where network service requirements are used as an input to generate an optimized abstract network inventory representation, called Logical Network Inventory (LNI). We also provide a formal definition of the network model and problem statement along with the development of three algorithms to efficiently build an LNI. Results show the potential benefits of using an LNI to streamline service management and placement: (i) the relationship between compute nodes and links (i.e., density) in an LNI is reduced between 1.8-2.7x compared to a full network inventory topology; and (ii) up to 50% of time can be saved for service placement after abstracting around 20% of the compute nodes.
{"title":"ANI: Abstracted Network Inventory for Streamlined Service Placement in Distributed Clouds","authors":"D. A. L. Perez, Christian Esteve Rothenberg, Mateus A. S. Santos, P. Gomes","doi":"10.1109/NetSoft48620.2020.9165443","DOIUrl":"https://doi.org/10.1109/NetSoft48620.2020.9165443","url":null,"abstract":"Scenarios for distributed cloud with multiple edge clouds and centralized data centers are being investigated as the computing and networking underpinnings of next-generation network services such as augmented reality, self-driving vehicles, drones, and more. In such distributed environments, service providers will typically face tens, hundreds, or thousands of compute location candidates (edge, regional, and central) where network service components can be placed. To take optimized placement decisions of network services and execute the management workflows, orchestration systems require up-to-date and accurate resource availability representation, in the form of a network inventory that can be immense in distributed cloud scenarios. As a result, the service management and placement problems may become not tractable. In this work, we propose the Abstracted Network Inventory (ANI) component to generate service-optimized network views over the same network inventory. ANI implements a novel abstraction method where network service requirements are used as an input to generate an optimized abstract network inventory representation, called Logical Network Inventory (LNI). We also provide a formal definition of the network model and problem statement along with the development of three algorithms to efficiently build an LNI. Results show the potential benefits of using an LNI to streamline service management and placement: (i) the relationship between compute nodes and links (i.e., density) in an LNI is reduced between 1.8-2.7x compared to a full network inventory topology; and (ii) up to 50% of time can be saved for service placement after abstracting around 20% of the compute nodes.","PeriodicalId":239961,"journal":{"name":"2020 6th IEEE Conference on Network Softwarization (NetSoft)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114703633","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 : 2020-06-01DOI: 10.1109/netsoft48620.2020.9165531
F. R. Cesen, Levente Csikor, C. Recalde, Christian Esteve Rothenberg, Gergely Pongrácz
Due to the advanced control and machine learning techniques, today's industrial robots are faster and more accurate than human workers in well-structured repetitive tasks. However, in case of sudden changes in the operational area, such as unexpected obstacles or humans, robots have to be continuously monitored by powerful controllers for swift interventions (i.e., send emergency stop signals). As in the case of many verticals (e.g., transportation, shopping), the proliferation of Software-Defined Networking (SDN) and Network Function Virtualization (NFV) has started to captivate industry 4.0 as well in order to benefit from low infrastructure costs, and flexible management and resource provisioning. Besides all the advantages of the centralized approach, however, in critical situations (e.g., possible collisions, actuator damages or human injuries) the required ultra-low latency between the robots and the controller becomes an all-important factor, and one of the main concerns, at the same time, for industry leaders making the decision towards this paradigm shift. In this paper, we argue that by relying on recently emerged stateful and programmable data planes, it is possible to fill this gap by offloading latency-critical applications to the network, thereby bringing some intelligence much closer the robots. We present the first in-network robotic control application that is capable to intercept the communication between the robot and the controller and craft responses immediately if needed. In particular, we show that we can detect position threshold violations entirely in the data plane, close to the robot, and deliver emergency stop commands within no time with full compliance to the actual TCP session and application states.
{"title":"Towards Low Latency Industrial Robot Control in Programmable Data Planes","authors":"F. R. Cesen, Levente Csikor, C. Recalde, Christian Esteve Rothenberg, Gergely Pongrácz","doi":"10.1109/netsoft48620.2020.9165531","DOIUrl":"https://doi.org/10.1109/netsoft48620.2020.9165531","url":null,"abstract":"Due to the advanced control and machine learning techniques, today's industrial robots are faster and more accurate than human workers in well-structured repetitive tasks. However, in case of sudden changes in the operational area, such as unexpected obstacles or humans, robots have to be continuously monitored by powerful controllers for swift interventions (i.e., send emergency stop signals). As in the case of many verticals (e.g., transportation, shopping), the proliferation of Software-Defined Networking (SDN) and Network Function Virtualization (NFV) has started to captivate industry 4.0 as well in order to benefit from low infrastructure costs, and flexible management and resource provisioning. Besides all the advantages of the centralized approach, however, in critical situations (e.g., possible collisions, actuator damages or human injuries) the required ultra-low latency between the robots and the controller becomes an all-important factor, and one of the main concerns, at the same time, for industry leaders making the decision towards this paradigm shift. In this paper, we argue that by relying on recently emerged stateful and programmable data planes, it is possible to fill this gap by offloading latency-critical applications to the network, thereby bringing some intelligence much closer the robots. We present the first in-network robotic control application that is capable to intercept the communication between the robot and the controller and craft responses immediately if needed. In particular, we show that we can detect position threshold violations entirely in the data plane, close to the robot, and deliver emergency stop commands within no time with full compliance to the actual TCP session and application states.","PeriodicalId":239961,"journal":{"name":"2020 6th IEEE Conference on Network Softwarization (NetSoft)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124920668","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 : 2020-06-01DOI: 10.1109/NetSoft48620.2020.9165522
D. Sanvito, A. Marchini, I. Filippini, A. Capone
Data-center topologies interconnect an ever larger number of servers using a high number of alternative paths to provide high bandwidth and a high degree of resiliency. The state-of-the-art routing strategy is based on Equal-cost multipath (ECMP) which employs static hashing mechanism over packet header fields to spread the traffic over multiple paths. Routing the traffic without considering the size of the flows and the utilization of the paths might cause congestion due to the collision of multiple large flows on a same downstream path. We present CEDRO, an in-switch mechanism to detect and reschedule colliding large flows. By exploiting the latest advances in SDN programmable network devices, we offload to the network the detection of both the elephant flows and the path congestion conditions and the rescheduling mechanism. CEDRO is able to promptly cope with path congestion and failures directly from the dataplane, regardless of the availability of the external controller. We implemented CEDRO in an emulated SDN network and tested it against realistic traffic scenarios. Numerical evaluation shows CEDRO is able to improve the average and 95-th percentile of the Flow Completion Time compared to ECMP.
{"title":"CEDRO: an in-switch elephant flows rescheduling scheme for data-centers","authors":"D. Sanvito, A. Marchini, I. Filippini, A. Capone","doi":"10.1109/NetSoft48620.2020.9165522","DOIUrl":"https://doi.org/10.1109/NetSoft48620.2020.9165522","url":null,"abstract":"Data-center topologies interconnect an ever larger number of servers using a high number of alternative paths to provide high bandwidth and a high degree of resiliency. The state-of-the-art routing strategy is based on Equal-cost multipath (ECMP) which employs static hashing mechanism over packet header fields to spread the traffic over multiple paths. Routing the traffic without considering the size of the flows and the utilization of the paths might cause congestion due to the collision of multiple large flows on a same downstream path. We present CEDRO, an in-switch mechanism to detect and reschedule colliding large flows. By exploiting the latest advances in SDN programmable network devices, we offload to the network the detection of both the elephant flows and the path congestion conditions and the rescheduling mechanism. CEDRO is able to promptly cope with path congestion and failures directly from the dataplane, regardless of the availability of the external controller. We implemented CEDRO in an emulated SDN network and tested it against realistic traffic scenarios. Numerical evaluation shows CEDRO is able to improve the average and 95-th percentile of the Flow Completion Time compared to ECMP.","PeriodicalId":239961,"journal":{"name":"2020 6th IEEE Conference on Network Softwarization (NetSoft)","volume":"116 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124622731","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：