FAST TCP

FAST TCP is a new TCP congestion avoidance algorithm especially targeted at high-speed, long-distance links, developed at the [http://netlab.caltech.edu/ Netlab] , California Institute of Technology and now being commercialized by [http://www.fastsoft.com Fastsoft] . It is compatible with existing TCP algorithms, requiring modification only to the computer which is sending data.

Name

The name FAST is a recursive acronym for FAST AQM Scalable TCP, where AQM stands for Active Queue Management, and TCP stands for Transmission Control Protocol.

Principles of operation

The role of congestion control is to moderate the rate at which data is transmitted, according to the capacity of the network and the rate at which other users are transmitting. Like TCP Vegas, FAST TCP [cite journal
last = Wei
first = David X.
coauthors = Jin, Cheng; Low, Steven H. and Hegde, Sanjay
date =
year = 2006
month =
title = FAST TCP: motivation, architecture, algorithms, performance
journal = IEEE/ACM Trans. on Networking
volume = 14
issue = 6
pages = 1246–1259
url = http://netlab.caltech.edu/pub/papers/FAST-ToN-final-060209.pdf
doi = 10.1109/TNET.2006.886335] [cite journal
last = Jin
first = Cheng
coauthors = Wei, David X.; Low, Steven H.; Buhrmaster, G.; Bunn, J.; Choe, D. H.; Cottrell, R. L. A.; Doyle, J. C.; Feng, W.; Martin, O.; Newman, H. Paganini, F.; Ravot, S.; Singh, S.
date =
year = 2005
month =
title = FAST TCP: from theory to experiments
journal = IEEE Network
volume = 19
issue = 1
pages = 4–11
url = http://netlab.caltech.edu/pub/papers/fast-network05.pdf
doi = 10.1109/MNET.2005.1383434 ] uses queueing delay instead of loss probability as a congestion signal.

Most current congestion control algorithms detect congestion and slow down when they discover that packets are being dropped, so that the average sending rate depends on the loss probability. This has two drawbacks. First, low loss probabilities are required to sustain high data rates; in the case of TCP Reno, very low loss probabilities are required, but even new congestion avoidance algorithms such as H-TCP, BIC TCP and HSTCP require loss rates lower than those provided by most wireless wide area networks. Moreover, packet loss only provides a single bit of information about the congestion level, whereas delay is a continuous quantity and in principle provides more information about the network.

A FAST TCP flow seeks to maintain a constant number of packets in queues throughout the network. The number of packets in queues is estimated by measuring the difference between the observed round trip time (RTT) and the "base RTT", defined as the round trip time when there is no queueing. The base RTT is estimated as the minimum observed RTT for the connection. If too few packets are queued, the sending rate is increased, while if too many are queued, the rate is decreased. In this respect, it is a direct descendant of TCP Vegas.

The difference between TCP Vegas and FAST TCP lies in the way in which the rate is adjusted when the number of packets stored is too small or large. TCP Vegas makes fixed size adjustments to the rate, independent of how far the current rate is from the target rate. FAST TCP makes larger steps when the system is further from equilibrium and smaller steps near equilibrium. This improves the speed of convergence and the stability.

trengths and weaknesses

Delay-based algorithms can, in principle, maintain a constant window size, avoiding the oscillations inherent in loss-based algorithms. However, they also detect congestion earlier than loss-based algorithms, since delay corresponds to partially filled buffers, while loss results from totally filled buffers. This can be either a strength or a weakness. If the only protocol used in a network is delay-based, then the inefficiency of loss can be avoided; however, if loss-based and delay-based protocols share the network [cite journal
last = Tang
first = Ao
coauthors = Wang, Jiantao; Low, Steven H. and Chiang, Mung
date =
year = 2005
month = March
title = Network Equilibrium of heterogeneous congestion control protocols
conference = IEEE INFOCOM
conferenceurl = http://www.ieee-infocom.org/2005/
location = Miami, FL
url = http://netlab.caltech.edu/pub/papers/multiprotocol-infocom05.pdf ] , then delay-based algorithms tend to be less aggressive. This can be overcome by suitable choice of parameters, leading to complex interactions studied by Tang et al.

Delay measurements are also subject to jitter as a result of operating system scheduling, or bus contention.

Whether the strengths or weaknesses prevail is not clear, and depends in large part on the particular scenario.

Intellectual property

Unlike most TCP congestion avoidance algorithms, FAST TCP is protected by several patents [cite web
last = Jin
first = Cheng
coauthors = Low, Steven H.; Wei, Xiaoliang
title = Method and apparatus for network congestion control
publisher = United States Patent & Trademark Office
date = 2005-01-27
url = http://appft1.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&p=1&u=%2Fnetahtml%2FPTO%2Fsearch-bool.html&r=2&f=G&l=50&co1=AND&d=PG01&s1=jin.IN.&s2=low.IN.&OS=IN/jin+AND+IN/low&RS=IN/jin+AND+IN/low
accessdate = 2006-11-05 ] [cite web
last = Jin
first = Cheng
coauthors = Low, Steven H.; Wei, Xiaoliang
coauthors = Low, Steven H.; Wei, David X.; Wydrowski, Bartek; Tang, Ao; Choe, Hyojeong
title = Method and apparatus for network congestion control using queue control and one-way delay measurements
publisher = United States Patent & Trademark Office
date = 2006-03-09
url = http://appft1.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&p=1&u=%2Fnetahtml%2FPTO%2Fsearch-bool.html&r=1&f=G&l=50&co1=AND&d=PG01&s1=jin.IN.&s2=low.IN.&OS=IN/jin+AND+IN/low&RS=IN/jin+AND+IN/low
accessdate = 2006-11-05 ] . Instead of seeking standardization by the IETF, the inventors of FAST, notably Steven Low and Cheng Jin, are seeking to commercialize it through the company [http://www.fastsoft.com FastSoft] . Currently FastSoft sells a 1-Unit rack appliance which can be deployed at the sender-side with no software or hardware modifications needed on either end.

References

ee also

* TCP congestion avoidance algorithm
* Transmission Control Protocol#Development_of_TCP

External links

* [http://netlab.caltech.edu/FAST/ FAST] Home Page.
* [http://ultralight.caltech.edu/web-site/sc05/html/index.html Supercomputing 2005 Bandwidth Challenge]