Wyniki wyszukiwania - Biblioteka Nauki

1

Modeling TCP/RED - approximate performance evaluation using an analytical queueing approach

100%

Świderski J.

Systems Science

|

2006

|

tom Vol. 32, no 3

109-117

EN

This paper is concerned with the approximate performance analysis of TCP Reno connections coexisting with flows that are not TCP-friendly (i.e. flows that use more bandwidth than would any conformant TCP implementation in the same circumstances). The network model is based on two earlier models. The first model enables one to calculate the TCP send rate which is the amount of data sent by the sender in unit time. This model has been presented in [6] and captures the behavior of both the fast retransmit and time-out mechanisms. The send rate for a given TCP connection can be calculated assuming that the round trip time, the time-out and the loss probability are known for this connection. These quantities can be determined using the second model – the queueing network model that represents an IP network and has been introduced in [13]. It enables one to include the Random Early Detection (RED) queue based active queue management (AQM) scheme that can be met in today’s routers.

2

Modelowanie dynamiki przesyłów TCP/IP z uwzględnieniem mechanizmu RED

89%

Nycz M. , Czachórski T.

|

tom Vol. 33, nr 3A

49-62

PL

W pracy przedstawiono wyniki modelowania wybranych scenariuszy współczesnych sieci komputerowych. W badaniach zastosowano model matematyczny aproksymacji fluid flow, opisujący wpływ zmiennego w czasie natężenia ruchu pakietów na przełączniki sieciowe i generowanie strat pakietowych, zgodnie z założeniami mechanizmu RED stosowanego w routerach . Przeprowadzone doświadczenia obejmowały wpływ początkowych ustawień parametrów na zachowanie układu.

EN

The paper presents the results of modeling of selected modern computer networks scenarios. The study used a mathematical model of fluid flow approxmiation, describing the effects of time varying packet traffic on the network switches and packet loss generation by the assumptions of the RED mechanism used in routers. The carried out esperiments included the influence of tlie parameters initial settings on the behavior of a whole system.

3

A model of TCP flows dynamics

88%

Czachórski T. , Grochla K. , Pekergin F.

|

tom Vol. 18, nr 4

293-310

EN

The fluid-flow approximation models iuvestigate with much success the dynamics and stability of TCP/RED connections. Their main assumption is that the fluctuations of variables characterizing the behaviour of the couuectioiis are relatively small, that enables the linearization of model and the use of traditional control analysis tools to obtain such measures as Bode gain, phase margins, tracking error or delay margin. The coutribution of this article is twofold: (i) preserving linear fluid-flow model, we propose its extension to the case when a network is composed of wired and wireless part. In this case we consider a variant of TCP algorithiu (TCP-DCR) and fluid-flow differential equations representing the size of congestion window. mean queue at the bottleneck router and loss probability at a RED queue are supplemented with terms representing constant loss probability due to transmission in wireless pait and probability that a fraction of these errors is recovered by a liuk levei mechanism. The decrease of congestion window due to TCP mechanism is delayed to allow the link protocol to deal with the errors. The nonlinear model is linearized and may be used to check the stability of closed-loop control. (ii) We propose to replace the linear fluid flow model of the congested router queue and its AQM politias by more complex one, based on diffusiou approximation. The model will allow us to follow the queue evolution and its dynamics as well as the dynamics of congestion control, also when the transmission is far from the working point, including the case of saturated buffer. It might be especially useful when the counection dynamics is investigated in presence of large uncontrolled flows of UDP.

PL

Modele oparte na aproksymacji płynnej pozwalają na analizy funcjonowania połączeń TCP w obecności aktywnych metod zarządzania kolejką, takich jak RED. Wykorzystując lineralizacje modelu aproksymacji płynnej wokół punktu pracy, można dokonać badania stabilności połączenia TCP wykreślając wykresy Bode'go oraz Nyquista. W artykule rozwinięto dwa zagadnienia: (i) Zaproponowano rozszerzenie modelu aproksymacji płynnej znanego z Literatury o przypadek sieci zbudowanej z 2 części - przewodowej i bezprzewodowej. Autorzy rozpatrują wariant algorytmu regulacji rozmiaru okna protokołu TCP o nazwie TCP-DCR, dla którego wyznaczono równania różniczkowe aproksymacji płytnnej opisujące rozmiar okna TCP. długość kolejki oraz prawdopodobieństwo strat pakietów w koejce RED, uwzględniające straty pakietów w bezprzewodowej części sieci. Opracowany model jest następnie zlineralizowarry w celu przeprowadzenia analizy stabilności, (ii) Autorzy proponują zastąpić liniowy model przeciążonego rutera modelem opartym na aproksymacji dyfuzyjnej, pozwalającym na bad zmian długości kolejki w obecności zewnętrznych źródeł ruchu np. UDP.

4

Sared: self-adaptive active queue management scheme for improving quality of service in network systems

75%

Adamu A. , Surajo Y. , Jafar M. T.

Computer Science

|

2021

|

tom T. 22 (2)

251–266

EN

Considering the phenomenal growth of network systems, congestion remains a threat to the quality of the service provided in such systems; hence, research on congestion control is still relevant. The Internet research community regards active queue management (AQM) as an effective approach for addressing congestion in network systems. Most of the existing AQM schemes possess static drop patterns and lack a self-adaptation mechanism; as such they do not work well for networks where the traffic load fluctuates. This paper proposes a self-adaptive random early detection (SARED) scheme that smartly adapts its drop pattern based on a current network’s traffic load in order to maintain improved and stable performance. Under light- to moderate-load conditions, SARED operates in nonlinear modes in order to maximize utilization and throughput, while it switches to a linear mode in order to avoid forced drops and congestion under high-load conditions. Our conducted experiments revealed that SARED provides optimal performance regardless of the condition of the traffic load.