The Experts below are selected from a list of 62100 Experts worldwide ranked by ideXlab platform

Sally Floyd - One of the best experts on this subject based on the ideXlab platform.

Isaac Keslassy - One of the best experts on this subject based on the ideXlab platform.

  • SIGCOMM - Virtualized Congestion Control
    Proceedings of the 2016 ACM SIGCOMM Conference, 2016
    Co-Authors: Bryce Cronkite-ratcliff, Aran Bergman, Madhusudhan Ravi, Ittai Abraham, Shay Vargaftik, Nick Mckeown, Isaac Keslassy
    Abstract:

    New Congestion Control algorithms are rapidly improving datacenters by reducing latency, overcoming incast, increasing throughput and improving fairness. Ideally, the operating system in every server and virtual machine is updated to support new Congestion Control algorithms. However, legacy applications often cannot be upgraded to a new operating system version, which means the advances are off-limits to them. Worse, as we show, legacy applications can be squeezed out, which in the worst case prevents the entire network from adopting new algorithms. Our goal is to make it easy to deploy new and improved Congestion Control algorithms into multitenant datacenters, without having to worry about TCP-friendliness with non-participating virtual machines. This paper presents a solution we call virtualized Congestion Control. The datacenter owner may introduce a new Congestion Control algorithm in the hypervisors. Internally, the hypervisors translate between the new Congestion Control algorithm and the old legacy Congestion Control, allowing legacy applications to enjoy the benefits of the new algorithm. We have implemented proof-of-concept systems for virtualized Congestion Control in the Linux kernel and in VMware’s ESXi hypervisor, achieving improved fairness, performance, and Control over guest bandwidth allocations.

  • Virtualized Congestion Control
    Proceedings of the 2016 conference on ACM SIGCOMM 2016 Conference - SIGCOMM '16, 2016
    Co-Authors: Bryce Cronkite-ratcliff, Aran Bergman, Madhusudhan Ravi, Ittai Abraham, Shay Vargaftik, Nick Mckeown, Isaac Keslassy
    Abstract:

    New Congestion Control algorithms are rapidly improving datacenters by reducing latency, overcoming incast, increas-ing throughput and improving fairness. Ideally, the operat-ing system in every server and virtual machine is updated to support new Congestion Control algorithms. However, legacy applications often cannot be upgraded to a new op-erating system version, which means the advances are off-limits to them. Worse, as we show, legacy applications can be squeezed out, which in the worst case prevents the entire network from adopting new algorithms. Our goal is to make it easy to deploy new and improved Congestion Control algorithms into multitenant datacenters, without having to worry about TCP-friendliness with non-participating virtual machines. This paper presents a solu-tion we call virtualized Congestion Control. The datacen-ter owner may introduce a new Congestion Control algo-rithm in the hypervisors. Internally, the hypervisors trans-late between the new Congestion Control algorithm and the old legacy Congestion Control, allowing legacy applications to enjoy the benefits of the new algorithm. We have imple-mented proof-of-concept systems for virtualized Congestion Control in the Linux kernel and in VMware's ESXi hypervi-sor, achieving improved fairness, performance, and Control over guest bandwidth allocations.

Eddie Kohler - One of the best experts on this subject based on the ideXlab platform.

Wu Hang - One of the best experts on this subject based on the ideXlab platform.

  • The Summary of Congestion Control Algorithm in Network
    Computer Science, 2007
    Co-Authors: Wu Hang
    Abstract:

    With the rapid development of Internet,various applications based on Internet have emerged and the data flows on Internet are increased abruptly.Network Congestion has become a serious problem and Congestion Control has been the focus of attention in the field of network.This paper emphatically describes typical algorithms in TCP and IP Congestion Control and some current important Congestion Control algorithms and points out advantages and disadvantages of these algorithms.Subsequently the drawback in designing Congestion Control algorithm at present is analyzed,and an interesting direction is provided in the end.

Bryce Cronkite-ratcliff - One of the best experts on this subject based on the ideXlab platform.

  • SIGCOMM - Virtualized Congestion Control
    Proceedings of the 2016 ACM SIGCOMM Conference, 2016
    Co-Authors: Bryce Cronkite-ratcliff, Aran Bergman, Madhusudhan Ravi, Ittai Abraham, Shay Vargaftik, Nick Mckeown, Isaac Keslassy
    Abstract:

    New Congestion Control algorithms are rapidly improving datacenters by reducing latency, overcoming incast, increasing throughput and improving fairness. Ideally, the operating system in every server and virtual machine is updated to support new Congestion Control algorithms. However, legacy applications often cannot be upgraded to a new operating system version, which means the advances are off-limits to them. Worse, as we show, legacy applications can be squeezed out, which in the worst case prevents the entire network from adopting new algorithms. Our goal is to make it easy to deploy new and improved Congestion Control algorithms into multitenant datacenters, without having to worry about TCP-friendliness with non-participating virtual machines. This paper presents a solution we call virtualized Congestion Control. The datacenter owner may introduce a new Congestion Control algorithm in the hypervisors. Internally, the hypervisors translate between the new Congestion Control algorithm and the old legacy Congestion Control, allowing legacy applications to enjoy the benefits of the new algorithm. We have implemented proof-of-concept systems for virtualized Congestion Control in the Linux kernel and in VMware’s ESXi hypervisor, achieving improved fairness, performance, and Control over guest bandwidth allocations.

  • Virtualized Congestion Control
    Proceedings of the 2016 conference on ACM SIGCOMM 2016 Conference - SIGCOMM '16, 2016
    Co-Authors: Bryce Cronkite-ratcliff, Aran Bergman, Madhusudhan Ravi, Ittai Abraham, Shay Vargaftik, Nick Mckeown, Isaac Keslassy
    Abstract:

    New Congestion Control algorithms are rapidly improving datacenters by reducing latency, overcoming incast, increas-ing throughput and improving fairness. Ideally, the operat-ing system in every server and virtual machine is updated to support new Congestion Control algorithms. However, legacy applications often cannot be upgraded to a new op-erating system version, which means the advances are off-limits to them. Worse, as we show, legacy applications can be squeezed out, which in the worst case prevents the entire network from adopting new algorithms. Our goal is to make it easy to deploy new and improved Congestion Control algorithms into multitenant datacenters, without having to worry about TCP-friendliness with non-participating virtual machines. This paper presents a solu-tion we call virtualized Congestion Control. The datacen-ter owner may introduce a new Congestion Control algo-rithm in the hypervisors. Internally, the hypervisors trans-late between the new Congestion Control algorithm and the old legacy Congestion Control, allowing legacy applications to enjoy the benefits of the new algorithm. We have imple-mented proof-of-concept systems for virtualized Congestion Control in the Linux kernel and in VMware's ESXi hypervi-sor, achieving improved fairness, performance, and Control over guest bandwidth allocations.