The Experts below are selected from a list of 2460 Experts worldwide ranked by ideXlab platform

Yanhui Geng - One of the best experts on this subject based on the ideXlab platform.

  • Luopan: Sampling-Based Load Balancing in Data Center Networks
    IEEE Transactions on Parallel and Distributed Systems, 2019
    Co-Authors: Peng Wang, George Trimponias, Yanhui Geng
    Abstract:

    Data center networks demand high-performance, robust, and practical data plane load balancing protocols. Despite progress, existing work falls short of meeting these requirements. We design, analyze, and evaluate Luopan, a novel sampling based load balancing protocol that overcomes these challenges. Luopan operates at flowcell granularity similar to Presto. It periodically samples a few paths for each Destination Switch and directs flowcells to the least congested one. By being congestion-aware, Luopan improves flow completion time (FCT), and is more robust to topological asymmetries compared to Presto. The sampling approach simplifies the protocol and makes it much more scalable for implementation in large-scale networks compared to existing congestion-aware schemes. We provide analysis to show that Luopan's periodic sampling has the same asymptotic behavior as instantaneous sampling: taking 2 random samples provides exponential improvements over 1 sample. We conduct comprehensive packet-level simulations with production workloads. The results show that Luopan consistently outperforms state-of-the-art schemes in large-scale topologies. Compared to Presto, Luopan with 2 samples improves the 99.9%ile FCT of mice flows by up to 35 percent, and average FCT of medium and elephant flows by up to 30 percent. Luopan also performs significantly better than Local Sampling with large asymmetry.

  • ICNP - Luopan: Sampling based load balancing in data center networks
    2016 IEEE 24th International Conference on Network Protocols (ICNP), 2016
    Co-Authors: Peng Wang, George Trimponias, Hongyuan Liu, Yanhui Geng
    Abstract:

    Data center networks demand high-performance, robust, and practical data plane load balancing protocols. Despite progress, existing work falls short of satisfying these requirements. We design and evaluate Luopan, a novel sampling based load balancing protocol that overcomes these challenges. Luopan operates at flowcell granularity similar to Presto. It periodically samples a few paths to each Destination Switch and directs flowcells to the least congested one. By being congestion-aware, Luopan improves flow completion time (FCT), and is more robust to topological asymmetries compared to Presto. The sampling approach simplifies the protocol and makes it much more scalable for implementation in large-scale networks compared to existing congestion-aware schemes. We conduct comprehensive packet-level simulations with a production workload. The results show that Luopan consistently outperforms state-of-the-art schemes in large-scale symmetric and asymmetric topologies. Compared to Presto, Luopan with 2 samples improves the 99%ile FCT of mice flows by up to 45%, and average FCT of medium flows by ∼20%.

Christian Cseh - One of the best experts on this subject based on the ideXlab platform.

  • Modeling and Simulation of an ABR Flow Control Algorithm Using a Virtual Source/Virtual Destination Switch
    SIMULATION, 2002
    Co-Authors: Christian Cseh
    Abstract:

    The Available Bit Rate (ABR) service class of Asynchronous Transfer Mode networks uses a feedback control mechanism to adapt to varying link capacities. The Virtual Source/Virtual Destination (VS/VD) technique offers the possibility to segment the otherwise end-to-end ABR control loop into separate loops. The improved feedback delay and the control of ABR traffic inside closed segments provide a better performance for ABR connections. This article will present the use of classical linear control theory to model and develop an ABR VS/VD flow control algorithm. Discrete event simulations are used to analyze the behavior of the algorithm with respect to transient behavior and correctness of the control model. Linear control theory offers the means to derive correct choices of parameters and to assess performance issues, like stability of the system, during the design phase. The performance goals are a high link utilization, fair bandwidth distribution, and robust operation in various environments, which are ...

  • modeling and simulation of an abr flow control algorithm using a virtual source virtual Destination Switch
    International Conference on Advances in System Simulation, 2002
    Co-Authors: Christian Cseh
    Abstract:

    The Available Bit Rate (ABR) service class of Asynchronous Transfer Mode networks uses a feedback control mechanism to adapt to varying link capacities. The Virtual Source/Virtual Destination (VS/VD) technique offers the possibility to segment the otherwise end-to-end ABR control loop into separate loops. The improved feedback delay and the control of ABR traffic inside closed segments provide a better performance for ABR connections. This article will present the use of classical linear control theory to model and develop an ABR VS/VD flow control algorithm. Discrete event simulations are used to analyze the behavior of the algorithm with respect to transient behavior and correctness of the control model. Linear control theory offers the means to derive correct choices of parameters and to assess performance issues, like stability of the system, during the design phase. The performance goals are a high link utilization, fair bandwidth distribution, and robust operation in various environments, which are ...

  • modeling and analysis of an abr flow control algorithm for a virtual source virtual Destination Switch
    Modeling Analysis and Simulation On Computer and Telecommunication Systems, 2001
    Co-Authors: Christian Cseh
    Abstract:

    The ABR service class of ATM networks uses a feedback control mechanism to adapt to varying link capacities. The VS/VD technique offers the possibility to segment the otherwise end-to-end ABR control loop into separate loops. The improved feedback delay and the control of ABR traffic inside closed segments provide a better performance for ABR connections. This paper presents the use of classical linear control theory to model and develop an ABR VS/VD flow control algorithm. Discrete event simulation is used to analyze the behavior of the algorithm with respect to transient behavior and correctness of the control model. Linear control theory offers the means to derive correct choices of parameters and to assess performance issues, like stability of the system, during the design phase. The performance goals are a high link utilization, fair bandwidth distribution and robust operation in various environments, which are verified by discrete event simulations. The major contribution of this work is the use of analytic methods (linear control theory) to model and design an ABR flow control algorithm tailored for the special layout of a VS/VD Switch, and the use of simulation techniques to verify the result.

  • MASCOTS - Modeling and analysis of an ABR flow control algorithm for a virtual source/virtual Destination Switch
    MASCOTS 2001 Proceedings Ninth International Symposium on Modeling Analysis and Simulation of Computer and Telecommunication Systems, 2001
    Co-Authors: Christian Cseh
    Abstract:

    The ABR service class of ATM networks uses a feedback control mechanism to adapt to varying link capacities. The VS/VD technique offers the possibility to segment the otherwise end-to-end ABR control loop into separate loops. The improved feedback delay and the control of ABR traffic inside closed segments provide a better performance for ABR connections. This paper presents the use of classical linear control theory to model and develop an ABR VS/VD flow control algorithm. Discrete event simulation is used to analyze the behavior of the algorithm with respect to transient behavior and correctness of the control model. Linear control theory offers the means to derive correct choices of parameters and to assess performance issues, like stability of the system, during the design phase. The performance goals are a high link utilization, fair bandwidth distribution and robust operation in various environments, which are verified by discrete event simulations. The major contribution of this work is the use of analytic methods (linear control theory) to model and design an ABR flow control algorithm tailored for the special layout of a VS/VD Switch, and the use of simulation techniques to verify the result.

Peng Wang - One of the best experts on this subject based on the ideXlab platform.

  • Luopan: Sampling-Based Load Balancing in Data Center Networks
    IEEE Transactions on Parallel and Distributed Systems, 2019
    Co-Authors: Peng Wang, George Trimponias, Yanhui Geng
    Abstract:

    Data center networks demand high-performance, robust, and practical data plane load balancing protocols. Despite progress, existing work falls short of meeting these requirements. We design, analyze, and evaluate Luopan, a novel sampling based load balancing protocol that overcomes these challenges. Luopan operates at flowcell granularity similar to Presto. It periodically samples a few paths for each Destination Switch and directs flowcells to the least congested one. By being congestion-aware, Luopan improves flow completion time (FCT), and is more robust to topological asymmetries compared to Presto. The sampling approach simplifies the protocol and makes it much more scalable for implementation in large-scale networks compared to existing congestion-aware schemes. We provide analysis to show that Luopan's periodic sampling has the same asymptotic behavior as instantaneous sampling: taking 2 random samples provides exponential improvements over 1 sample. We conduct comprehensive packet-level simulations with production workloads. The results show that Luopan consistently outperforms state-of-the-art schemes in large-scale topologies. Compared to Presto, Luopan with 2 samples improves the 99.9%ile FCT of mice flows by up to 35 percent, and average FCT of medium and elephant flows by up to 30 percent. Luopan also performs significantly better than Local Sampling with large asymmetry.

  • ICNP - Luopan: Sampling based load balancing in data center networks
    2016 IEEE 24th International Conference on Network Protocols (ICNP), 2016
    Co-Authors: Peng Wang, George Trimponias, Hongyuan Liu, Yanhui Geng
    Abstract:

    Data center networks demand high-performance, robust, and practical data plane load balancing protocols. Despite progress, existing work falls short of satisfying these requirements. We design and evaluate Luopan, a novel sampling based load balancing protocol that overcomes these challenges. Luopan operates at flowcell granularity similar to Presto. It periodically samples a few paths to each Destination Switch and directs flowcells to the least congested one. By being congestion-aware, Luopan improves flow completion time (FCT), and is more robust to topological asymmetries compared to Presto. The sampling approach simplifies the protocol and makes it much more scalable for implementation in large-scale networks compared to existing congestion-aware schemes. We conduct comprehensive packet-level simulations with a production workload. The results show that Luopan consistently outperforms state-of-the-art schemes in large-scale symmetric and asymmetric topologies. Compared to Presto, Luopan with 2 samples improves the 99%ile FCT of mice flows by up to 45%, and average FCT of medium flows by ∼20%.

George Trimponias - One of the best experts on this subject based on the ideXlab platform.

  • Luopan: Sampling-Based Load Balancing in Data Center Networks
    IEEE Transactions on Parallel and Distributed Systems, 2019
    Co-Authors: Peng Wang, George Trimponias, Yanhui Geng
    Abstract:

    Data center networks demand high-performance, robust, and practical data plane load balancing protocols. Despite progress, existing work falls short of meeting these requirements. We design, analyze, and evaluate Luopan, a novel sampling based load balancing protocol that overcomes these challenges. Luopan operates at flowcell granularity similar to Presto. It periodically samples a few paths for each Destination Switch and directs flowcells to the least congested one. By being congestion-aware, Luopan improves flow completion time (FCT), and is more robust to topological asymmetries compared to Presto. The sampling approach simplifies the protocol and makes it much more scalable for implementation in large-scale networks compared to existing congestion-aware schemes. We provide analysis to show that Luopan's periodic sampling has the same asymptotic behavior as instantaneous sampling: taking 2 random samples provides exponential improvements over 1 sample. We conduct comprehensive packet-level simulations with production workloads. The results show that Luopan consistently outperforms state-of-the-art schemes in large-scale topologies. Compared to Presto, Luopan with 2 samples improves the 99.9%ile FCT of mice flows by up to 35 percent, and average FCT of medium and elephant flows by up to 30 percent. Luopan also performs significantly better than Local Sampling with large asymmetry.

  • ICNP - Luopan: Sampling based load balancing in data center networks
    2016 IEEE 24th International Conference on Network Protocols (ICNP), 2016
    Co-Authors: Peng Wang, George Trimponias, Hongyuan Liu, Yanhui Geng
    Abstract:

    Data center networks demand high-performance, robust, and practical data plane load balancing protocols. Despite progress, existing work falls short of satisfying these requirements. We design and evaluate Luopan, a novel sampling based load balancing protocol that overcomes these challenges. Luopan operates at flowcell granularity similar to Presto. It periodically samples a few paths to each Destination Switch and directs flowcells to the least congested one. By being congestion-aware, Luopan improves flow completion time (FCT), and is more robust to topological asymmetries compared to Presto. The sampling approach simplifies the protocol and makes it much more scalable for implementation in large-scale networks compared to existing congestion-aware schemes. We conduct comprehensive packet-level simulations with a production workload. The results show that Luopan consistently outperforms state-of-the-art schemes in large-scale symmetric and asymmetric topologies. Compared to Presto, Luopan with 2 samples improves the 99%ile FCT of mice flows by up to 45%, and average FCT of medium flows by ∼20%.

Hongyuan Liu - One of the best experts on this subject based on the ideXlab platform.

  • ICNP - Luopan: Sampling based load balancing in data center networks
    2016 IEEE 24th International Conference on Network Protocols (ICNP), 2016
    Co-Authors: Peng Wang, George Trimponias, Hongyuan Liu, Yanhui Geng
    Abstract:

    Data center networks demand high-performance, robust, and practical data plane load balancing protocols. Despite progress, existing work falls short of satisfying these requirements. We design and evaluate Luopan, a novel sampling based load balancing protocol that overcomes these challenges. Luopan operates at flowcell granularity similar to Presto. It periodically samples a few paths to each Destination Switch and directs flowcells to the least congested one. By being congestion-aware, Luopan improves flow completion time (FCT), and is more robust to topological asymmetries compared to Presto. The sampling approach simplifies the protocol and makes it much more scalable for implementation in large-scale networks compared to existing congestion-aware schemes. We conduct comprehensive packet-level simulations with a production workload. The results show that Luopan consistently outperforms state-of-the-art schemes in large-scale symmetric and asymmetric topologies. Compared to Presto, Luopan with 2 samples improves the 99%ile FCT of mice flows by up to 45%, and average FCT of medium flows by ∼20%.