Improving performance of delay-based TCPs with rerouting (original) (raw)
Related papers
Delay-based congestion avoidance for TCP
IEEE/ACM Transactions on Networking, 2003
The set of TCP congestion control algorithms associated with TCP/Reno (e.g., slow-start and congestion avoidance) have been crucial to ensuring the stability of the Internet. Algorithms such as TCP/NewReno (which has been deployed) and TCP/Vegas (which has not been deployed) represent incrementally deployable enhancements to TCP as they have been shown to improve a TCP connection's throughput without degrading performance to competing flows. Our research focuses on delay-based congestion avoidance algorithms (DCA), like TCP/Vegas, which attempt to utilize the congestion information contained in packet round-trip time (RTT) samples. Through measurement and simulation, we show evidence suggesting that a single deployment of DCA (i.e., a TCP connection enhanced with a DCA algorithm) is not a viable enhancement to TCP over high-speed paths. We define several performance metrics that quantify the level of correlation between packet loss and RTT. Based on our measurement analysis we find that although there is useful congestion information contained within RTT samples, the level of correlation between an increase in RTT and packet loss is not strong enough to allow a TCP/Sender to reliably improve throughput. While DCA is able to reduce the packet loss rate experienced by a connection, in its attempts to avoid packet loss, the algorithm will react unnecessarily to RTT variation that is not associated with packet loss. The result is degraded throughput as compared to a similar flow that does not support DCA.
VFAST TCP: A delay-based enhanced version of FAST TCP
This paper is aimed at describing a delay-based endto-end (e2e) congestion control algorithm, called Very FAST TCP (VFAST), which is an enhanced version of FAST TCP. The main idea behind this enhancement is to smoothly estimate the Round-Trip Time (RTT) based on a nonlinear filter, which eliminates throughput and queue oscillation when RTT fluctuates. In this context, an evaluation of the suggested scheme through simulation is introduced, by comparing our VFAST prototype with FAST in terms of throughput, queue behavior, fairness, stability, RTT and adaptivity to changes in network. The achieved simulation results indicate that the suggested protocol offer better performance than FAST TCP in terms of RTT estimation and throughput.
An enhancement scheme for TCP over mobile ad hoc networks
The 57th IEEE Semiannual Vehicular Technology Conference, 2003. VTC 2003-Spring., 2003
In a Mobile Ad Hoc Network, temporary link failures and route changes occur frequently. With the assumption that all packet losses are due to congestion, TCP performs poorly in such an environment. This paper proposes a new mechanism called TSR, TCP-aware Source Routing, which can improve TCP performance in wireless ad hoc networks. TSR adds a hold state to an existing routing protocol to reduce consecutive timeouts, retransmissions, and out-of-ordered packets in TCP. In our simulation study, TSR achieves up to a 60% improvement in performance, without requiring any TCP stacks in end systems to be modified.
Internet Research Task Force (irtf) Delay-tolerant Networking Tcp Convergence-layer Protocol
This document describes the protocol for the TCP-based convergence layer for Delay-Tolerant Networking (DTN). It is the product of the IRTF's DTN Research Group (DTNRG). Task Force (IRTF). Documents approved for publication by the IRSG are not a candidate for any level of Internet Standard; see Section 2 of RFC 5741. Information about the current status of this document, any errata, and how to provide feedback on it may be obtained at
International Journal of Communication Systems, 2015
SummaryNetwork utilization by legacy transmission control protocol (TCP) is determined by its round trip time (RTT) dependent mechanism for the growth of its sending rate. The RTT does not always reflect the actual network conditions, especially in the case of wireless local area network (WLAN). Consequently, it influences the RTT‐dependent mechanism falsely. This paper proposes a novel cross‐layer scheme between TCP and the Institute of Electrical and Electronics Engineers (IEEE) 802.11 medium access control (MAC) that compensates for any diminished growth of TCP's sending rate because of the inclusion of non‐congestion delay component in RTT, if any. The proposed scheme has two refinements. The first refinement is at the MAC layer that notifies the additional propagation delay on account of link retransmission of a TCP packet. The second refinement is at the TCP layer in which the sender adapts the sending rate by relating the aforementioned additional propagation delay to the...
Improving TCP Performance by Packet Buffering in Mobile IP Based Networks
2004
It is well-known that TCP often experiences severe performance degradation in mobile networks since due to host mobility, packet losses not related to network congestion occur frequently. In this paper, we propose a buffering of packets at a base station to resolve such a problem. Our method can be used without sacrificing the scalability in Mobile IP based networks. For this purpose, we first investigate the performance of TCP without considering packet buffering through simulation experiments. Our results show that in most cases, the smooth handoff by the route optimization extension of the Mobile IP standard cannot prevent degradation of TCP performance in evens of handoffs, although it was originally intended to reduce the number of packets dropped during the handoff. It is also shown that in utilizing the route optimization extension, the TCP performance sometimes becomes worse even than the case of the base Mobile IP, if its smooth handoff fails to avoid losses of four or more packets during the handoff. Such results indicate that for TCP, the smooth handoff is not useful unless the route optimization extension supports a buffering method, which makes handoffs be transparent to transport layer protocols by recovering lost packets during the handoff. We next investigate the impacts of packet buffering on TCP performance. We modify the route optimization extension in order to support packet buffering at the base station, which only requires minor changes. Finally, we discuss some problems occurring when recovering the packets dropped during the handoff by the buffering method, and propose our solution.
DVPTCP: A Delay-driven Virtual Parallel TCP for High-speed and Lossy Networks
IEEE Access, 2019
The growing diversification of network architectures and application demands results in poor performance of most existing TCPs, especially in large bandwidth-delay product (BDP) and lossy networks. To tackle this problem, we propose a novel delay-based congestion control algorithm, named DVPTCP, which estimates the network congestion level by using round-trip time (RTT) and adjusts the number of virtual parallel streams dynamically. Related theoretical analyses about throughput and TCP-friendliness are given in detail. According to our extensive simulation tests, our proposal can achieve better bandwidth utilization and TCP-friendliness than the compared schemes in large-BDP and lossy networks.
Performance Enhancement of TCP in Mobile IP Based Networks
2012
In mobile IP based networks, most of the packet drops occur due to mobile handoff from one sub-network to another sub-network or larger delays in between sender and receiver due to encapsulation and tunnelling of packets. Packet loss during the handoff operation in mobile IP with route optimization and smooth handover will degrade the performance of Transmission Control Protocol (TCP). To prevent performance degradation due to mobile handover packet drops, a number of packet buffering and packet forwarding mechanisms have been proposed. However, mobile user continually change its point-of-attachment and can sometimes move into congested base station of foreign agent sub network, its buffered forwarded packets are likely to be dropped at the new base station. This can lead to decrease in congestion window size at the sender TCP, which will severely degrade the performance of TCP even though there is no congestion within the network. In this approach, hardware resources (packet buffers) are efficiently utilized, at the same time performance degradation of TCP due to congested base station of current mobile node is minimized through packet buffers, markers and with modification in Mobile IP binding update message.
Computer Communications, 2007
Packet losses during the handoff operation by the route optimization extension of the Mobile IP causes performance degradation at the transmission control protocol (TCP). To prevent such degradation a number of packet buffering based methods have been proposed in the literature. However, as the mobile host user continually changes location and can sometimes move into a congested BS in a new foreign subnetwork, its buffered packets are likely to be dropped at the new BS. This can lead to losses at the TCP connections of the mobile user host in the new subnetwork, as well as at the TCP connections of the new BS which experiences severe performance degradation due to the abrupt increase in congestion by the forwarded burst of packets (i.e., global synchronization). This paper proposes a priority forwarding (PF) scheme designed to significantly improve the performance of the packet buffering methods. The proposed PF scheme does not require any modification to the Mobile IP protocol with route optimization extension. The simulation results show that by using the PF scheme the period of global synchronization at the inter-subnetwork handoff is made shorter due to the reduction in packet droppings in the random early detection (RED) buffer. This improves the TCP performance in wireless networks employing Mobile IP with packet buffering.