FAST TCP: Motivation, Architecture, Algorithms, Performance

By David X. Wei, Cheng Jin, Steven H. Low, Sanjay Hegde. IEEE/ACM Transactions on Networking, 2006

1,413 citations recorded by the source corpus.

Subfields: Computer Networks and Communications, Electrical and Electronic Engineering

View Paper: https://www.semanticscholar.org/paper/14e35dff0929e56200a56375faa730b40edb97f4

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

CUBIC: a new TCP-friendly high-speed TCP variant (2008, background): “To counter this under-utilization problem of TCP, many “high-speed” TCP variants are proposed (e.g., FAST [24], HSTCP [15], STCP [25], HTCP [28], SQRT [19], West-wood [14], and BIC-TCP [30]).”
Performance Isolation and Fairness for Multi-Tenant Cloud Storage (2012, positive): “It thus adjusts the request windows according to the FAST-TCP [34] update: w(m + 1) t n = (1−α) · w(m) t n + α · l base l est Each iteration of the algorithm adjusts the window based on how close the tenant demand is to its local rate allocation, which is represented by the ratio of a desired averag”
Copa: Practical Delay-Based Congestion Control for the Internet (2018, methodology): “FAST [42] derives its equilibrium properties from utility maximization.”
Stochastic Forecasts Achieve High Throughput and Low Delay over Cellular Networks (2013, neutral): “TCP Vegas [4], FAST [12], and Compound TCP [20] incorporate round-trip delay explicitly, but the adaptation is reactive and does not directly involve the receiver’s observed rate.”
Sizing router buffers (2004, background): “TCP flavors that use latency (and not drops) to detect congestion [8, 26] have very different buffer requirements and are based on a fixed amount of buffering needed per flow, independent of the delay-bandwidth-product.”
Data center TCP (DCTCP) (2010, background): “Much research has been devoted, with success, to improving TCP performance on paths with high bandwidth-delay product, including High-speed TCP [8], CUBIC [30] and FAST [33].”
H-TCP : TCP for high-speed and long-distance networks (2004, methodology): “Protocols for high-speed and long distance networks Recently, the design of congestion control protocols for deployment in high speed and long distance networks has been the subject of much interest in the networking community [7] [8] [9].”
Less Is More: Trading a Little Bandwidth for Ultra-Low Latency in the Data Center (2012, methodology): “Within the TCP family, variants like Vegas [11], CTCP [48] and FAST [53] attempt to control congestion by reacting to increasing RTTs due to queuing delay.”
TIMELY: RTT-based Congestion Control for the Datacenter (2015, methodology): “Delay has been explored in the wide-area Internet since at least TCP Vegas [16], and some modern TCP variants use delay estimates [44, 46].”
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].”
Network Optimization and Control (2008, background): “in [63], and implementation details of FAST TCP are provided in [64].”
Real-Time Convex Optimization in Signal Processing (2010, methodology): “It is widely used in application areas like control [6], [7], [8], circuit design [9], [10], [11], economics and finan ce [12], [13], networking [14], [15], [16], statistics and mac hine learning [17], [18], quantum information theory [19], [20] , and combinatorial optimization [21], to name just a ”
Host-to-Host Congestion Control for TCP (2010, background): “[82]–[84], inspired by the Vegas idea of congestion control with the queuing delay as a primary congestion indicator (see Section II-G), introduced a FAST algorithm.”
The Globus Striped GridFTP Framework and Server (2005, methodology): “These solutions include improvements to TCP [19, 33, 37], new transport protocols such as XCP [36], XTP [52] and reliable layers on top of UDP [5, 13, 14, 27, 30, 33, 51].”