Understanding TCP over TCP: effects of TCP tunneling on end-to-end throughput and latency

@inproceedings{Honda2005UnderstandingTO,
  title={Understanding TCP over TCP: effects of TCP tunneling on end-to-end throughput and latency},
  author={Osamu Honda and Hiroyuki Ohsaki and Makoto Imase and Mika Ishizuka and Junichi Murayama},
  booktitle={SPIE Optics East},
  year={2005}
}
TCP tunnel is a technology that aggregates and transfers packets sent between end hosts as a single TCP connection. By using a TCP tunnel, the fairness among aggregated flows can be improved and several protocols can be transparently transmitted through a firewall. Currently, many applications such as SSH, VTun, and HTun use a TCP tunnel. However, since most applications running on end hosts generally use TCP, two TCP congestion controls (i.e., end-to-end TCP and tunnel TCP) operate… 
Effects of TCP Transfer Buffers and Congestion Avoidance Algorithms on the End-to-End Throughput of TCP-over-TCP Tunnels
TLDR
Through ns-3 simulation, this research has investigated the contributions that multiple factors play in degrading the goodput of TCP-over-TCP tunnels, and found the performance of 448 different configurations, not counting experimental control cases.
Multipath Transport for Virtual Private Networks
TLDR
This paper empirically quantify the potential of using MPTCP tunnels to increase the goodput of VPN communications when multiple data paths are available and design and implement a preliminary version of Multipath UDP (MPUDP) to address the adverse effect of the duplicated congestion control actions that is known with a TCP-in-TCP tunnel.
Throughput modeling and analysis for TCP over TCP satellite communications
TLDR
A throughput system model is developed, and an analysis of the impacts of TCP retransmission is provided, which shows a throughput reduction when both inner and outer TCP react to packet loss.
Benefits of Multiply-Cascaded TCP Tunnel with Network Coding over Lossy Networks
TLDR
To utilize the potential of TCP/NC more practically with no change on end-host TCP, this paper considers the TCP/ NC tunnel that simply conveys end-to-end TCP sessions not only on a single TCP/ND session but also on cascaded TCP/N sessions traversing a lossy network in the middle without per-session management.
Analysing TCP performance when link experiencing packet loss
TLDR
Experimental results show that the congestion control algorithm H-TCP performed better than Cubic and Reno while link was experiencing packet loss, and ACK loss didn't affect performance that much, and up to 50% loss of ACKs could be tolerated with almost no performance degradation.
Masking lossy networks by TCP tunnel with Network Coding
TLDR
The results show that the proposed TCP/NC tunnel can mitigate the goodput degradation of end-to-end TCP sessions traversing a lossy network without any change in TCP on each end host.
SSH over UDP
TLDR
A UDP based solution performs slightly better than a TCP based solution on links experiencing packet loss, and is modified by adding support for a UDP base connection to OpenSSH's VPN functionality.
Understanding the Effects of TCP Tunneling on Approaching High Performance HAIPE-Encrypted Satellite Communication Systems
TLDR
The potential solution of using TCP tunnel for SATCOM links with high error rate is evaluated to provide a better understanding of the TCP tunneling performance and to identify and develop effective solutions for improving TCP throughput over encrypted SATCOM networks.
T3: TCP-Based High-Performance and Congestion-Aware Tunneling Protocol for Cloud Networking
TLDR
This paper proposes a novel tunneling protocol named "Transparent Transport Tunneling (T3)" that realizes high-performance communications with traditional network devices and supports a congestion control mechanism and results showed that T3 achieved STT comparable throughput as well as 1/20 average queue length in a middlebox device.
Deploying Internet Protocol Security in satellite networks using Transmission Control Protocol Performance Enhancing Proxies
TLDR
Three different approaches are proposed to deploy IPsec in a scenario with TCP PEPs, which provide different trade‐offs between security and TCP performance in some typical scenarios that use satellite networks.
...
...

References

SHOWING 1-10 OF 15 REFERENCES
Effectiveness of TCP SACK, TCP HACK and TCP Trunk over satellite links
  • L. Jacob, S. Nair, Huang Duo, A. Ananda
  • Computer Science
    2002 IEEE International Conference on Communications. Conference Proceedings. ICC 2002 (Cat. No.02CH37333)
  • 2002
This paper reports a study on the performance enhancements of two extensions to the standard TCP implementation - Selective Acknowledgement (SACK) and Header Checksum (HACK) - over satellite links
TCP trunking: design, implementation and performance
TLDR
TC trunking can be an effective tool for network operators in managing bandwidth sharing between aggregates and can guarantee minimum bandwidth while being able to grab additional bandwidth when it is available.
On individual and aggregate TCP performance
  • L. Qiu, Yin Zhang, S. Keshav
  • Business, Computer Science
    Proceedings. Seventh International Conference on Network Protocols
  • 1999
TLDR
This paper develops a simple yet realistic network model to abstract an Internet connection and studies the performance of a single TCP flow with many competing TCP flows by evaluating the best-known analytical model proposed in the literature.
The macroscopic behavior of the TCP congestion avoidance algorithm
TLDR
A performance model for the TCP Congestion Avoidance algorithm that predicts the bandwidth of a sustained TCP connection subjected to light to moderate packet losses, such as loss caused by network congestion is analyzed.
A stochastic model of TCP/IP with stationary random losses
TLDR
A model for TCP/IP congestion control mechanism is presented and an explicit expression for the throughput of a TCP connection is obtained and bounds on the throughput when there is a limit on the window size.
A Comparison of Equation-Based and AIMD Congestion Control
TLDR
This paper considers AIMD-based congestion control mechanisms that are TCP-compatible, but that reduce their sending rate less sharply than does TCP in response to a single packet drop, and compares them with TFRC (TCP-Friendly Rate Control), which makes use of equation- based congestion control.
SoftEther による Ethernet の仮想化とトンネリング通 信
  • 情報処理学会 第 45 回プログラミングシンポジウム, Jan. 2004.
  • 2004
HTun: Providing IP service over an HTTP proxy
  • HTun: Providing IP service over an HTTP proxy
HTun : Providing IP service over an HTTP proxy
    ...
    ...