Equation-based congestion control for unicast applications

By Sally Floyd, Mark James Handley, J. Padhye, J. Widmer. Conference on Applications, Technologies, Architectures, and Protocols for Computer Communication, 2000

1,689 citations recorded by the source corpus.

This paper proposes a mechanism for equation-based congestion control for unicast traffic. Most best-effort traffic in the current Internet is well-served by the dominant transport protocol, TCP. However, traffic such as best-effort unicast streaming multimedia could find use for a TCP-friendly congestion control mechanism that refrains from reducing the sending rate in half in response to a single packet drop. With our mechanism, the sender explicitly adjusts its sending rate as a function of the measured rate of loss events, where a loss event consists of one or more packets dropped within a single round-trip time. We use both simulations and experiments over the Internet to explore performance.

Subfields: Computer Networks and Communications, Electrical and Electronic Engineering

View Paper: https://www.semanticscholar.org/paper/ebb492dbd9c524b96f8711e657c1d6672d361c7e

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Networks and Communications

In Electrical and Electronic Engineering

Citations in the corpus

PRIME: Peer-to-Peer Receiver-drIven MEsh-Based Streaming (2007, methodology): “In summary, when peer degree is large, an aggressive congestion control mechanism such as RAP may cause a rather higher loss rate and thus content bottleneck whereas slower congestion control mechanisms such as TFRC reduce the loss rate at the cost of lower utilization of resources.”
Congestion control for high bandwidth-delay product networks (2002, methodology): “This failing becomes increasingly important as the Internet evolves to incorporate very high-bandwidth optical links and more large-delay satellite links.”
FAST TCP: Motivation, Architecture, Algorithms, Performance (2006, methodology): “Without explicit feedback, can only be loss probability, as used in TFRC [13], or queueing delay, as used in TCP Vegas [3] and FAST TCP. Queueing delay can be more accurately estimated than loss probability both because loss samples provide coarser information than queueing delay samples, and…”
pgmcc: a TCP-friendly single-rate multicast congestion control scheme (2000, methodology): “We also plan to investigate, as future work, the techniques used in TFRC [12] for measuring losses.”
Internet Protocol Television (IPTV): The Killer Application for the Next-Generation Internet (2007, methodology): “Even though TFRC [13] tries to smooth out rate variability, it does not take the properties of the overlying media application into consideration when protecting media flows from short-term congestion.”
Bullet: high bandwidth data dissemination using an overlay mesh (2003, methodology): “First, it must be TCP friendly [15].”
Rate-distortion optimized streaming of packetized media (2006, methodology): “In this approach, the congestion control mechanism makes a local estimate of TCP's long-term average transmission rate using an equation for the number of seconds between data units such as [68] T = Rp2 R=3+3( R+4 R) R(1+32 2R)p3 R=8; (16) where R, R, and 2 R are short-term estimates of the packet l”
A survey on TCP-friendly congestion control (2001, methodology): “Similar to TFRCP it adjusts its sending rate based on the complex TCP equation (2) but uses more sophisticated methods to gather the necessary parameters.”
General AIMD congestion control (2000, background): “The congestion control schemes investigated can be divided into two categories: AIMD-based [13, 24, 4, 23, 19] and formula-based [17, 29, 30, 26, 21, 10].”
Characterizing residential broadband networks (2007, methodology): “Traditionally, transport protocols have been developed to achieve stable throughput and to avoid reacting to short-term events (on timescales less than one RTT) [19].”
TCP ex machina: computer-generated congestion control (2013, background): “Other schemes developed in the literature include equationbased congestion control [16], binomial control [6], FastTCP [42], HSTCP, and TCP Westwood [30].”
Finding a "Kneedle" in a Haystack: Detecting Knee Points in System Behavior (2011, background): “Likewise, since multiple flows share network links in the Internet, network protocols cannot predict in advance the rapidly changing level of TCP-friendly bandwidth available, but must instead continuously adapt to the indirect signals of packet loss and delay [6].”
End-to-end differentiation of congestion and wireless losses (2001, methodology): “For congestion control, we use the TCP-Friendly Rate Control (TFRC) algorithm [4].”
Extending equation-based congestion control to multicast applications (2001, background): “TFRC is an equation-based congestion control scheme.”
H.264/AVC over IP (2003, background): “Media rate control schemes that take TCP-friendliness into account have been reported in academia [8], but have not yet seen wide deployment in the field—mostly because, so far, no one on the public Internet is “enforcing” the TCP traffic paradigm.”