Two schemes to reduce latency in short lived TCP flows

Authors:
D. Ciullo;M. Mellia;M. Meo
Affiliations:
Dipartimento di Elettronica, Politecnico di Torino, Italy;Dipartimento di Elettronica, Politecnico di Torino, Italy;Dipartimento di Elettronica, Politecnico di Torino, Italy
Venue:
IEEE Communications Letters
Year:
2009

Citing 3
Cited 4

Wide area traffic: the failure of Poisson modeling

IEEE/ACM Transactions on Networking (TON)
TCP smart framing: a segmentation algorithm to reduce TCP latency

IEEE/ACM Transactions on Networking (TON)
Measuring IP and TCP behavior on edge nodes with Tstat

Computer Networks: The International Journal of Computer and Telecommunications Networking

A new TCP mechanism for reducing retransmission timeouts over multi-hop wireless networks

Proceedings of the 2011 ACM Symposium on Applied Computing
Required extra capacity: A comparative estimation of overprovisioning needed for a classless IP backbone

Computer Networks: The International Journal of Computer and Telecommunications Networking
On the trade-off between spectrum efficiency with dedicated access and short end-to-end transmission delays with random access in DVB-RCS2

Proceedings of the 2013 ACM MobiCom workshop on Lowest cost denominator networking for universal access
FavorQueue: A parameterless active queue management to improve TCP traffic performance

Computer Networks: The International Journal of Computer and Telecommunications Networking

Quantified Score

Hi-index	0.00

Visualization

Abstract

Motivated by the amount of short-lived TCP traffic due to interactive applications like web browsing, we propose two schemes to reduce the download time of TCP flows. Both schemes explicitly try to overcome the long delay suffered when TCP has to recover from the loss of the last data segment in a flow, loss for which Fast Retransmit cannot be triggered. The first scheme consists in transmitting twice the last segment, while the second one exploits the reception of the duplicate ACK caused by the delivery of the FIN segment. Both schemes allow the transmitter to recover the loss without waiting for a Retransmission Timeout expiration, thus providing quicker recovery with up to 50% gain.