The Experts below are selected from a list of 7773 Experts worldwide ranked by ideXlab platform

Panagiotis Papadimitriou - One of the best experts on this subject based on the ideXlab platform.

  • 5G World Forum - Leveraging on Source Routing for Scalability and Robustness in Datacenters
    2019 IEEE 2nd 5G World Forum (5GWF), 2019
    Co-Authors: Konstantinos Papadopoulos, Panagiotis Papadimitriou
    Abstract:

    The transition towards 5G leads to higher in-network processing demands, which, in turn, raise more severe dataplane requirements, e.g., in terms of datacenter switch Forwarding State. Since legacy switches coupled with layer-2/3 Forwarding are expected to introduce significant dataplane limitations, we leverage on source routing as means for a more scalable and robust switching fabric, especially for datacenters with large-scale network function deployments. In this respect, we present a practical implementation of source routing using OpenFlow-enabled switches. To assess the gains of source routing in virtualized infrastructures, we perform a comparative study between source routing and L2 Forwarding, using our implementation in Mininet. Our evaluation results show significant Forwarding State savings and hassle-free network updates, when the proposed source routing scheme is employed.

  • CNSM - Towards Reduced-State Service Chaining with Source Routing
    2018
    Co-Authors: Chrysa Papagianni, Panagiotis Papadimitriou, John S. Baras
    Abstract:

    The widespread adoption of Network Function Virtualization (NFV) poses significant dataplane scalability limitations in datacenters (DC), as switches will be required to maintain a large amount of Forwarding State for service chaining. This problem is exacerbated by the small Forwarding Information Base (FIB) of commodity switches, which are typically deployed in NFV infrastructures.To mitigate this problem, we present a routing fabric that combines source routing with pathlet switching in order to (i) achieve State reduction and (ii) provide support for longer paths, hence, meeting the increased hop-count requirement of service chains. Coupling the proposed source routing fabric with a service chain embedding method, we achieve significant gains in terms of FIB consumption, request acceptance, and revenue generation.

  • WWIC - SDN-Based Source Routing for Scalable Service Chaining in Datacenters
    Lecture Notes in Computer Science, 2016
    Co-Authors: Ahmed Abujoda, Hadi Razzaghi Kouchaksaraei, Panagiotis Papadimitriou
    Abstract:

    The migration of network functions (NFs) to datacenters, as promoted by Network Function Virtualization (NFV), raises the need for service chaining (i.e., steering traffic through a sequence of NFs). Service chaining is typically performed by installing Forwarding entries in switches within datacenters (DCs). However, as the number of service chains in DCs grows, switches will be required to maintain a large amount of Forwarding State. This will raise a dataplane scalability issue, due to the relatively small flow table size of switches. To mitigate this problem, we present a software-defined network (SDN) based source routing architecture for scalable service chaining, at which the NF-path is encoded into the packet header obviating the need for any Forwarding State and lookup in the switches. We assess the feasibility and efficiency of our architecture using a prototype implementation.

  • SDN-based source routing for scalable service chaining in datacenters
    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2016
    Co-Authors: Ahmed Abujoda, Hadi Razzaghi Kouchaksaraei, Panagiotis Papadimitriou
    Abstract:

    The migration of network functions (NFs) to datacenters, as promoted by Network Function Virtualization (NFV), raises the need for service chaining (i.e., steering trafficthroughasequence of NFs). Service chaining is typically performed by installing Forwarding entries in switches within datacenters (DCs). However, as the number of service chains in DCs grows, switches will be required tomaintain a large amount of Forwarding State. This will raise a dataplane scalability issue, due to the relatively small flow table size of switches. To mitigate this problem, we present a software-defined network (SDN) based source routing archi- tecture for scalable service chaining, at which the NF-path is encoded into the packet header obviating the need for any Forwarding State and lookup in the switches. We assess the feasibility and efficiency of our architecture using a prototype implementatio

Ahmed Abujoda - One of the best experts on this subject based on the ideXlab platform.

  • WWIC - SDN-Based Source Routing for Scalable Service Chaining in Datacenters
    Lecture Notes in Computer Science, 2016
    Co-Authors: Ahmed Abujoda, Hadi Razzaghi Kouchaksaraei, Panagiotis Papadimitriou
    Abstract:

    The migration of network functions (NFs) to datacenters, as promoted by Network Function Virtualization (NFV), raises the need for service chaining (i.e., steering traffic through a sequence of NFs). Service chaining is typically performed by installing Forwarding entries in switches within datacenters (DCs). However, as the number of service chains in DCs grows, switches will be required to maintain a large amount of Forwarding State. This will raise a dataplane scalability issue, due to the relatively small flow table size of switches. To mitigate this problem, we present a software-defined network (SDN) based source routing architecture for scalable service chaining, at which the NF-path is encoded into the packet header obviating the need for any Forwarding State and lookup in the switches. We assess the feasibility and efficiency of our architecture using a prototype implementation.

  • SDN-based source routing for scalable service chaining in datacenters
    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2016
    Co-Authors: Ahmed Abujoda, Hadi Razzaghi Kouchaksaraei, Panagiotis Papadimitriou
    Abstract:

    The migration of network functions (NFs) to datacenters, as promoted by Network Function Virtualization (NFV), raises the need for service chaining (i.e., steering trafficthroughasequence of NFs). Service chaining is typically performed by installing Forwarding entries in switches within datacenters (DCs). However, as the number of service chains in DCs grows, switches will be required tomaintain a large amount of Forwarding State. This will raise a dataplane scalability issue, due to the relatively small flow table size of switches. To mitigate this problem, we present a software-defined network (SDN) based source routing archi- tecture for scalable service chaining, at which the NF-path is encoded into the packet header obviating the need for any Forwarding State and lookup in the switches. We assess the feasibility and efficiency of our architecture using a prototype implementatio

R K C Chang - One of the best experts on this subject based on the ideXlab platform.

  • Forwarding State scalability aware multicast routing
    International Symposium on Computers and Communications, 2005
    Co-Authors: R K C Chang
    Abstract:

    Multicast routing protocols today still scale poorly to a large number of concurrent multicast sessions in terms of Forwarding States. Unlike previous approaches which concentrated on reducing Forwarding States after constructing multicast trees, our approach is to make the underlying routing algorithms aware of the scalability requirement. This scalability-aware approach can be applied to many existing multicast State reduction methods, such as aggregated multicast (AM) and dynamic tunnel multicast (DTM). We have formulated both AM-aware and DTM-aware routing problems as multicriteria optimization problems, and proposed algorithms to solve them.

  • ISCC - Forwarding State scalability-aware multicast routing
    10th IEEE Symposium on Computers and Communications (ISCC'05), 1
    Co-Authors: R K C Chang
    Abstract:

    Multicast routing protocols today still scale poorly to a large number of concurrent multicast sessions in terms of Forwarding States. Unlike previous approaches which concentrated on reducing Forwarding States after constructing multicast trees, our approach is to make the underlying routing algorithms aware of the scalability requirement. This scalability-aware approach can be applied to many existing multicast State reduction methods, such as aggregated multicast (AM) and dynamic tunnel multicast (DTM). We have formulated both AM-aware and DTM-aware routing problems as multicriteria optimization problems, and proposed algorithms to solve them.

Hadi Razzaghi Kouchaksaraei - One of the best experts on this subject based on the ideXlab platform.

  • WWIC - SDN-Based Source Routing for Scalable Service Chaining in Datacenters
    Lecture Notes in Computer Science, 2016
    Co-Authors: Ahmed Abujoda, Hadi Razzaghi Kouchaksaraei, Panagiotis Papadimitriou
    Abstract:

    The migration of network functions (NFs) to datacenters, as promoted by Network Function Virtualization (NFV), raises the need for service chaining (i.e., steering traffic through a sequence of NFs). Service chaining is typically performed by installing Forwarding entries in switches within datacenters (DCs). However, as the number of service chains in DCs grows, switches will be required to maintain a large amount of Forwarding State. This will raise a dataplane scalability issue, due to the relatively small flow table size of switches. To mitigate this problem, we present a software-defined network (SDN) based source routing architecture for scalable service chaining, at which the NF-path is encoded into the packet header obviating the need for any Forwarding State and lookup in the switches. We assess the feasibility and efficiency of our architecture using a prototype implementation.

  • SDN-based source routing for scalable service chaining in datacenters
    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2016
    Co-Authors: Ahmed Abujoda, Hadi Razzaghi Kouchaksaraei, Panagiotis Papadimitriou
    Abstract:

    The migration of network functions (NFs) to datacenters, as promoted by Network Function Virtualization (NFV), raises the need for service chaining (i.e., steering trafficthroughasequence of NFs). Service chaining is typically performed by installing Forwarding entries in switches within datacenters (DCs). However, as the number of service chains in DCs grows, switches will be required tomaintain a large amount of Forwarding State. This will raise a dataplane scalability issue, due to the relatively small flow table size of switches. To mitigate this problem, we present a software-defined network (SDN) based source routing archi- tecture for scalable service chaining, at which the NF-path is encoded into the packet header obviating the need for any Forwarding State and lookup in the switches. We assess the feasibility and efficiency of our architecture using a prototype implementatio

Mario Gerla - One of the best experts on this subject based on the ideXlab platform.

  • Aggregated Multicast – A Comparative Study
    Cluster Computing, 2005
    Co-Authors: Jun-hong Cui, Jinkyu Kim, Dario Maggiorini, Khaled Boussetta, Mario Gerla
    Abstract:

    Though IP multicast is resource efficient in delivering data to a group of members simultaneously, it suffers from scalability problem with the number of concurrently active multicast groups because it requires a router to keep Forwarding State for every multicast tree passing through it. To solve this State scalability problem, we proposed a scheme, called aggregated multicast . The key idea is that multiple groups are forced to share a single delivery tree. In our earlier work, we introduced the basic concept of aggregated multicast and presented some initial results to show that multicast State can be reduced. In this paper, we develop a more quantitative assessment of the cost/benefit trade-offs. We propose an algorithm to assign multicast groups to delivery trees with controllable cost and introduce metrics to measure multicast State and tree management overhead for multicast schemes. We then compare aggregated multicast with conventional multicast schemes, such as source specific tree scheme and shared tree scheme. Our extensive simulations show that aggregated multicast can achieve significant routing State and tree management overhead reduction while containing the expense of extra resources (bandwidth waste and tunnelling overhead). We conclude that aggregated multicast is a very cost-effective and promising direction for scalable transit domain multicast provisioning.

  • Networked Group Communication - Aggregated Multicast with Inter-Group Tree Sharing
    Lecture Notes in Computer Science, 2001
    Co-Authors: Aiguo Fei, Jun-hong Cui, Mario Gerla, Michalis Faloutsos
    Abstract:

    IP multicast suffers from scalability problems for large numbers of multicast groups, since each router keeps Forwarding State proportional to the number of multicast tree passing through it. In this paper, we present and evaluate aggregated multicast, an approach to reduce multicast State. In aggregated multicast, multiple groups are forced to share a single delivery tree. At the expense of some bandwidth wastage, this approach can reduce multicast State and tree management overhead at transit routers. It may also simplify and facilitate the provisioning of QoS guarantee for multicast in future aggregated-flow-based QoS networks. We formulate the tree sharing problem and propose a simple intuitive algorithm. We study this algorithm and evaluate the trade-off of aggregation vs. bandwidth overhead using simulations. Simulation results show that significant aggregation is achieved while at the same time bandwidth overhead can be reasonably controlled.

  • IWDC - Aggregated Multicast for Scalable QoS Multicast Provisioning
    Lecture Notes in Computer Science, 2001
    Co-Authors: Mario Gerla, Aiguo Fei, Jun-hong Cui, Michalis Faloutsos
    Abstract:

    IP multicast suffers from scalability problem with the number of concurrently active multicast groups, while scalability of QoS multicast is even further from being solved. In this paper, we propose an approach to reduce multicast Forwarding State and provision multicast with QoS guarantees. In our approach, multiple groups are forced to share a single delivery tree. We discuss the advantages and some implementation issues of our approach, and conclude that it is feasible and promising. We then describe how to use our approach to provision scalable QoS multicast. Finally, we define metrics to quantify State reduction and use simulations to show how our scheme achieves State reduction. These initial simulation results suggest that our method can reduce multicast State significantly.

  • GLOBECOM - Aggregated multicast: an approach to reduce multicast State
    GLOBECOM'01. IEEE Global Telecommunications Conference (Cat. No.01CH37270), 1
    Co-Authors: Aiguo Fei, Jun-hong Cui, Mario Gerla, Michalis Faloutsos
    Abstract:

    IP multicast suffers from a scalability problem with the number of concurrently active multicast groups because it requires a router to keep the Forwarding State for every multicast tree passing through it and the number of Forwarding entries grows with the number of groups. In this paper, we propose an approach to reduce the multicast Forwarding State. In our approach, multiple groups are forced to share a single delivery tree. We discuss the advantages and some implementation issues of our approach, and conclude that it is feasible and promising. We then propose metrics to quantify State reduction and analyze the bounds on State reduction of our approach. Finally, we use simulations to verify our analytical bounds and quantify the State reduction. These initial simulation results suggest that our method can reduce multicast State significantly.

  • GLOBECOM - Scalable QoS multicast provisioning in Diff-Serv-supported MPLS networks
    Global Telecommunications Conference 2002. GLOBECOM '02. IEEE, 1
    Co-Authors: Jun-hong Cui, Aiguo Fei, Michalis Faloutsos, Jinkyu Kim, Mario Gerla
    Abstract:

    IP multicast suffers from scalability problems as the number of concurrent active multicast groups increases, since it requires a router to keep a Forwarding State for every multicast tree passing through it. In QoS multicast provisioning, the problem is exacerbated, since not only the Forwarding State but also the resource requirement of a multicast group must be kept at the router. To provide scalable QoS multicast support, in this paper, we propose a novel architecture, called Aggregated QoS Multicast (AQoSM). Using the concept of aggregated multicast, AQoSM can support QoS multicast scalably and efficiently in DiffServ-supported MPLS networks. In this paper, we develop the framework for the architecture and provide a feasibility check from an implementation point of view. The architecture is flexible and can be customized to the needs and the existing protocols of a domain. Our simulations indicate that the architecture performs well in several common scenarios. It achieves smaller blocking of users with strong QoS requirements because of its load balancing capability. It also achieves up to 85% reduction in State with a modest 10% of bandwidth overhead.