The window distribution of idealized TCP congestion avoidance with variable packet loss

Misra, Archan; Ott, Teunis J.

doi:10.1109/infcom.1999.752179

Cited by 42 publications

(50 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[25] showed that throughput decreases faster (roughly proportional to 1/p) at higher values of p. All these papers assume constant round-trip times. Misra and Ott [21] incorporated state-dependent packet loss probabilities into their model and studied the stationary distribution of the congestion window.…”

Section: Probabilistic Analysis Of Tcp Congestion Controlmentioning

confidence: 99%

TCP is competitive against a limited adversary

Edmonds

Datta

Dymond

2003

Proceedings of the Fifteenth Annual ACM Symposium on Parallel Algorithms and Architectures

View full text Add to dashboard Cite

While the well-known Transport Control Protocol (TCP) is a de facto standard for reliable communication on the Internet, and performs well in practice, the question "how good is the TCP/IP congestion control algorithm?" is not completely resolved. In this paper, we provide some answers to this question using the competitive analysis framework. First, we prove that for networks with a single bottleneck (or point of congestion), TCP is competitive to the optimal global algorithm in minimizing the user-perceived latency or flow time of the sessions. Specifically, we show that with O(1) times as much bandwidth and O(1) extra time per job, TCP is O(1)-competitive against an optimal global algorithm. We motivate the need for allowing TCP to have extra resources by observing that existing lower bounds for non-clairvoyant scheduling algorithms imply that no online, distributed, non-clairvoyant algorithm can be competitive with an optimal offline algorithm if both algorithms were given the same resources. Second, we show that TCP is fair by proving that it converges quickly to allocations where every session gets its fair share of network bandwidth.

show abstract

Section: Probabilistic Analysis Of Tcp Congestion Controlmentioning

confidence: 99%

TCP is competitive against a limited adversary

Edmonds

Datta

Dymond

2003

Proceedings of the Fifteenth Annual ACM Symposium on Parallel Algorithms and Architectures

View full text Add to dashboard Cite

show abstract

“…To derive the probability distribution (and hence, other statistics such as the variance) of the congestion window of a generalized TCP flow (for = 1 :0), we modified the approach described in [13], which presented an analyticalcum-numerical technique to compute the TCP window distribution when congestion was signaled via packets dropped with a state-dependent probability. A purely analytical computation of the window distribution of an idealized TCP flow subject to packet drops with a constant drop probability was presented in [12].…”

Section: A Motivation and Related Workmentioning

confidence: 99%

“…4 As in [13], our formulation can also be used when different flows have marking probabilities that are scalar multiples of each other, i.e., f i mark (Q) = i f mark (Q) where i are arbitrary constants. We do not explore this scenario in this paper.…”

Section: A Characterizing the Fixed Pointmentioning

confidence: 99%

“…We accordingly use the technique presented in [13] (which considered the special case of current TCP congestion avoidance). This approach uses a set of state-dependent mappings to define an associated process X(t), which can be characterized by a differential equation between Poisson points of failure.…”

Section: B Solving the Stochastic Processmentioning

confidence: 99%

“…We use a generalization of the technique presented in [13] to determine the stationary window distribution and other window-related statistics in this case. By studying how such statistics vary as a function of the parameters of the generalized congestion avoidance algorithm, we deduce the relative importance of various suggested modifications on the behavior of the TCP source.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Generalized TCP congestion avoidance and its effect on bandwidth sharing and variability

Misra¹,

Baras

Ott³

Globecom '00 - IEEE. Global Telecommunications Conference. Conference Record (Cat. No.00CH37137)

Self Cite

View full text Add to dashboard Cite

Abstract-To model possible suggested changes in TCP window adaptation in response to randomized feedback, such as ECN, we formulate a generalized version of the TCP congestion avoidance algorithm. We first consider multiple such generalized TCP flows sharing a bottleneck buffer under the Assured Service model and use a fixed point technique to obtain the mean window sizes and throughputs for the TCP flows. To further study how changes in the adaptation algorithm affect the variability in the throughput, we use an analytical-cum-numerical technique to derive the window distribution (and related statistics) of a single generalized flow under state-dependent randomized congestion feedback.

show abstract

On the Accuracy of Analytical Models of TCP Throughput

Khayat

Geurts

Leduc

2006

NETWORKING 2006. Networking Technologies, Services, and Protocols; Performance of Computer and Communication Networks; Mobile A

View full text Add to dashboard Cite

Based on a large set of TCP sessions we first study the accuracy of two well-known analytical models (SQRT and PFTK) of the TCP average rate. This study shows that these models are far from being accurate on average. Actually, our simulations show that 70% of their predictions exceed the boundaries of TCP-Friendliness, thus questioning their use in the design of new TCP-Friendly transport protocols. Our study also shows that the inaccuracy of the PFTK model is largely due to its inability to make the distinction between the two packet loss detection methods used by TCP: triple duplicate acknowledgments or timeout expirations. We then use supervised learning techniques to infer models of the TCP rate. These models show important accuracy improvements when they take into account the two types of losses. This suggests that analytical model of TCP throughput should certainly benefit from the incorporation of the timeout loss rate.

show abstract

The window distribution of idealized TCP congestion avoidance with variable packet loss

Cited by 42 publications

References 8 publications

TCP is competitive against a limited adversary

TCP is competitive against a limited adversary

Generalized TCP congestion avoidance and its effect on bandwidth sharing and variability

On the Accuracy of Analytical Models of TCP Throughput

Contact Info

Product

Resources

About