• Corpus ID: 10604992

A Step toward Realistic Performance Evaluation of High-Speed TCP Variants

@inproceedings{Ha2006AST,
  title={A Step toward Realistic Performance Evaluation of High-Speed TCP Variants},
  author={Sangtae Ha and Yusung Kim and Long Le and Injong Rhee and Lisong Xu},
  year={2006}
}
This is a work-in-progress report on our work on designing realistic evaluation suites for testing high-speed TCP variants. In this work, we have created an experimental network model that captures some of the complex characteristics of propagation delays and background traffic [14, 15, 20]. We use our network model to evaluate a large collection of recently proposed TCPs for high-speed networks: BIC TCP, CUBIC, FAST, HSTCP, H-TCP, and STCP. While we do not claim that we have the most realistic… 

Figures from this paper

A model-driven emulation approach to large-scale TCP performance evaluation
TLDR
This work developed scalable virtualised evaluation environment for TCP (SVEET), a TCP performance evaluation testbed where real implementations of TCP variants can be accurately evaluated under diverse network configurations and workloads from real applications in large-scale network settings.
Towards a User-Oriented Benchmark for Transport Protocols Comparison in very High Speed Networks
TLDR
The goal of this report is to present some steps towards a user-oriented benchmark, called ITB, for high speed transport protocols comparison, by identifying classes of representative applications and useful metrics and synthesising comprehensive and useful properties.
A comparison of TCP behaviour at high speeds using ns-2 and Linux
TLDR
This work compares with TCP NewReno the fairness of the congestion control algorithms for 5 high-speed TCP variants -- BIC, Cubic, Scalable, High-Speed and Hamilton -- on both ns-2 and on an experimental testbed running Linux.
On the Performance Evaluation of High-Speed Transport Protocols
TLDR
In a controlled environment, XCP gave much better performance than the other TCPs, but was sensitive to misconfiguration and environmental factors, though, and was more difficult to deploy.
SVEET! a scalable virtualized evaluation environment for TCP
  • M. Erazo, Yue Li, Jason Liu
  • Computer Science
    2009 5th International Conference on Testbeds and Research Infrastructures for the Development of Networks & Communities and Workshops
  • 2009
TLDR
This work proposes a TCP performance evaluation testbed, called SVEET, on which real implementations of the TCP variants can be accurately evaluated under diverse network configurations and workloads in large-scale network settings.
A comparative performance evaluation of DCCP
  • S. Bhatti, M. Bateman, D. Miras
  • Computer Science
    2008 International Symposium on Performance Evaluation of Computer and Telecommunication Systems
  • 2008
TLDR
This work assesses the performance of DCCP CCID2 relative to TCP NewReno, and variants BIC and CUBIC, and also assesses fairness - how well these protocols might respond to each other when operating over the same end-to-end network path.
An improved method for experimental evaluations of TCP Congestion Control Algorithms
TLDR
A custom made utility with more abilities than existing tools and a set of scripts for automating benchmarks and graphing collected data are presented.
Empirical Analysis of TCP Variants and Their Impact on GridFTP Port Requirements
TLDR
It is proved that alternatives to TCP Reno become applicable today, better performance is achievable at the cost of unfairness and operating systems relying on TCP Reno may become disadvantaged in the future with new TCP variants.
Experimental and Analytical Analysis of TCP and UDP Protocols
TLDR
Analysis of the performance of Transmission Control Protocol through simulation and modeling finds that TCP SACK consistently performs better than TCP Reno, while for networks with large bandwidth-delay product, HS-TCP outperforms both.
An extended evaluation of a collection of TCP Congestion Control Algorithms
TLDR
This study is an experimental comparison and evaluation of Reno,Veno, Highspeed, BIC, CUBIC, Westwood, H-TCP, Hybla,Vegas, Scalable and Low-Priority algorithms.
...
...

References

SHOWING 1-10 OF 28 REFERENCES
Experimental Evaluation of TCP Protocols for High-Speed Networks
TLDR
It is found that both scalable-TCP and FAST- TCP consistently exhibit substantial unfairness, even when competing flows share identical network path characteristics.
Scalable TCP: improving performance in highspeed wide area networks
TLDR
The preliminary results gathered suggest that the deployment of Scalable TCP would have negligible impact on existing network traffic at the same time as improving bulk transfer performance in highspeed wide area networks.
Improving TCP Performance in High Bandwidth High RTT Links Using Layered Congestion Control
TLDR
It is shown that LTCP has promising convergence properties, is about an order of magnitude faster than TCP in utilizing high bandwidth links, employs few parameters and is easy to understand.
Dummynet: a simple approach to the evaluation of network protocols
TLDR
D dummynet works by intercepting communications of the protocol layer under test and simulating the effects of finite queues, bandwidth limitations and communication delays, and runs in a fully operational system, hence allowing the use of real traffic generators and protocol implementations.
TCP-Africa: an adaptive and fair rapid increase rule for scalable TCP
TLDR
This work proposes a new delay sensitive-congestion avoidance mode (TCP-Africa) that allows for scalable, aggressive behavior in large underutilized links, yet falls back to the more conservative TCP-Reno algorithm once links become well utilized and congestion is imminent.
Time for a TCP Benchmark Suite ?
TLDR
The challenges in designing a benchmark suite for TCP congestion control algorithms are summarized, a straw man proposal is given, and some findings are listed with the proposed benchmark.
The effect of router buffer size on HighSpeed TCP performance
TLDR
An analytical model is derived and it is shown that for a small buffer size, equal to 10% of the bandwidth-delay product, HighSpeed TCP can achieve more than 90%" of the bottleneck capacity.
CUBIC: a new TCP-friendly high-speed TCP variant
TLDR
The CUBIC protocol modifies the linear window growth function of existing TCP standards to be a cubic function in order to improve the scalability of TCP over fast and long distance networks.
Observations on the dynamics of a congestion control algorithm: the effects of two-way traffic
TLDR
This paper uses simulation to study the dynamics of the congestion cent rol algorithm embedded in the BSD 4.3-Tahoe TCP implementation and finds that the one-way traffic clustering and loss-synchronization phenomena persist in this new situation, albeit in a slightly modified form.
Generating Realistic TCP Workloads
TLDR
This work proposes a new methodology for transforming anonymized traces of packet headers into application-neutral models of network traffic, suitable for synthetic traffic generation in simulations and testbeds, preserving the end-to-end nature ofnetwork traffic.
...
...