1 / 16

Hiroyuki Ohsaki Cybermedia Center, Osaka University, Japan oosaki@cmc.osaka-u.ac.jp

Analysis of a window-based flow control mechanism based on TCP Vegas in heterogeneous network environment. Hiroyuki Ohsaki Cybermedia Center, Osaka University, Japan oosaki@cmc.osaka-u.ac.jp. Contents. Introduction Analytic model Window-based flow control based on TCP Vegas

marvel
Télécharger la présentation

Hiroyuki Ohsaki Cybermedia Center, Osaka University, Japan oosaki@cmc.osaka-u.ac.jp

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Analysis of a window-based flow control mechanism based on TCP Vegas in heterogeneous network environment Hiroyuki Ohsaki Cybermedia Center, Osaka University, Japan oosaki@cmc.osaka-u.ac.jp

  2. Contents • Introduction • Analytic model • Window-based flow control based on TCP Vegas • Heterogeneous network • Analysis of throughput and fairness • Analysis of stability and transient behavior • Numerical examples • Two bottleneck links with 30 TCP connections • Conclusion

  3. TCP (Transmission Control Protocol) • Packet retransmission mechanism • Retransmit lost packets in the network • Congestion avoidance mechanism • A window-based flow control mechanism • Several versions of TCP • TCP Tahoe • TCP Reno • TCP Vegas

  4. TCP Vegas • Advantages over TCP Reno • A new retransmission mechanism • An improved congestion avoidance mechanism • A modified slow-start mechanism • Uses measured RTT as feedback information 1. Measures RTT for a specific packet 2. Estimates severity of congestion 3. Changes window size • Packet loss can be prevented

  5. Objectives • Analyze a window-based flow control • Congestion avoidance mechanism of TCP Vegas • Connections with different propagation delays • Several bottleneck links • Using a control theoretic approach • Show numerical examples • Throughput and fairness • Stability and transient behavior

  6. Congestion avoidance of TCP Vegas • Source host maintains the minimum RTT: t • Source host measures the actual RTT: r(k) • Window size is changed based on d(k)

  7. Assumptions • Network topology is arbitrary • All routers employ static routing • All routers have a FIFO buffer for each port • All TCP connections are symmetry • Backward path is never congested

  8. State transition equation: window size • Window size of connection c : wc dc: control parameter that determines the amount of increase/decrease in window size

  9. State transition equation: queue length • Queue length of the buffer for link l : ql • b(c,l) : the previous link of link l for connection c • lc: the access link of connection c

  10. Throughput and fairness • Focus on equilibrium values (denoted by *) • rc* : throughput of connection c • qc* : sum of all queuing delays of connection c • Can be regarded as a Little's law • Fairness between TCP connections c and c'

  11. Stability and transient behavior • Obtain a linearized model • x(t) : state vector (current state – equilibrium state) • DLCM : Lowest Common Multiple of all tc/t's • Eigenvalues of A determinestability and transient behavior

  12. Numerical example: network model • Two bottleneck links and 30 TCP connections C3 Source Host Router C1 N1 N2 N3 L1 L2 Destination Host C2

  13. 10 rc1* 9 rc2* 8 rc3* 7 6 Throughput (packet/ms) 5 4 3 2 1 0 0 20 40 60 80 100 Link capacity of l2 (packet/ms) Numerical example: throughput • Effect of difference in link capacities (ml1 = 20)

  14. Numerical example: stability • Effect of link capacity (ml = l1 = l2) Control parameter dc of connection C2, C3 7 ml = 0.2 6 ml = 2 5 ml = 20 ml = 200 4 ml = 2000 3 2 1 0 0 1 2 3 4 5 6 7 Control parameter dc of connection C1

  15. Conclusion • Analytic model • Window-based flow control based on TCP Vegas • Homogeneous network • Throughput and fairness • Can be explained by Little's law • Has a bias against link capacity and # of bottleneck links • Window size in steady state • (bandwidth) x (propagation delay) + (control parameter gc)

  16. Conclusion (cont.) • Stability and transient behavior • Determined by eigenvalues of state transition matrix A • Link capacity significantly affects stability • Investigation using trajectories of eigenvalues • Future work • More simulation studies • Extension to TCP Tahoe or TCP Reno

More Related