Challenging the square-root law for TCP send-rate

Zaragoza, Daniel

doi:10.1049/el:20062423

Cited by 1 publication

(1 citation statement)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Furthermore, according to [Baccelli and McDonald 2005], inversesquare root proportionality is also valid with non-persistent flows. Nevertheless, the inverse-square root law for TCP throughput prediction is not universal; if packet loss is sufficiently small, the steady-state TCP throughput prediction can be obtained with the proportion 1/p (linear regime) instead of 1/ √ p if p is not greater than a value of p max , as stated in [Zaragoza 2006]. Just as an additional example, regarding any TCP lossbased congestion control protocol, a more general formula is stated by [Xu et al 2004] in R(p) = 1 RT T c p y ; R(p) is the response function representing the steady-state sending rate of the protocol in the unit of packets per RT T , stated as a function of packet loss (p), being c and y constants depending on the implementation.…”

Section: Tcp Throughput Predicitionmentioning

confidence: 99%

TCPBP: Enhancing the TCP Throughput in High Speed Lossy Networks With a Single Mixed Routing Metric

Vilela

Fonseca²,

Fonseca³

2022

Anais Do XL Simpósio Brasileiro De Redes De Computadores E Sistemas Distribuídos (SBRC 2022)

View full text Add to dashboard Cite

In this work, we present Transmission Control Protocol Best Path (TCPBP), an alternative link-state routing metric aims to optimize single-path TCP throughput on intradomain High Speed Lossy Networks. The enhanced performance is achieved by combining delay and packet loss on a single mixed metric as a function parameter. In contrast, this function predicts the TCP throughput that should be maximized. As proof of concept, TCP bandwidth tests were executed in an SDN High Speed Lossy Network Linux-emulated topology with ten nodes, employing an intradomain link-state routing protocol (i.e., OSPF-like) capable of running both TCPBP and traditional metrics delay and hop-count. Such topology was chosen because of the potential to improve its TCP throughput with route selection, given the characteristics of the topology itself and the TCP throughput prediction model adopted. The overall TCP throughput performance obtained with TCPBP was greater than with different routes founded by traditional metrics, showing the potential of our proposal.

show abstract

Section: Tcp Throughput Predicitionmentioning

confidence: 99%