09-05-2011, 12:21 PM
ABSTRACT
TCP Smart Framing, or TCP-SF for short, enables the Fast etransmit/ Recovery algorithms even when the congestion window is small. Without modifying the TCP congestion control based on the additiveincrease/ multiplicative-decrease paradigm, TCP-SF adopts a novel segmentation algorithm: while Classic TCP always tries to send full-sized segments, a TCP-SF source adopts a more flexible segmentation algorithm to try and always have a number of in-flight segments larger than 3 so as to enable Fast Recovery. We motivate this choice by real traffic measurements, which indicate that today’s traffic is populated by short-lived flows, whose only means to recover from a packet loss is by triggering a Retransmission Timeout. The key idea of TCP-SF can be implemented on top of any TCP lavor, from Tahoe to SACK, and requires modifications to the server TCP stack only, and can be easily coupled with recent TCP enhancements. Analytical results, simulation results, as well as testbed implementations and measurements show that TCP-SF sources outperform Classic TCP in terms of Completion Time. Also, simulations show that, when sharing a common bottleneck, TCP-SF steals little or no bandwidth from Classic TCP.
TCP Smart Framing, or TCP-SF for short, enables the Fast etransmit/ Recovery algorithms even when the congestion window is small. Without modifying the TCP congestion control based on the additiveincrease/ multiplicative-decrease paradigm, TCP-SF adopts a novel segmentation algorithm: while Classic TCP always tries to send full-sized segments, a TCP-SF source adopts a more flexible segmentation algorithm to try and always have a number of in-flight segments larger than 3 so as to enable Fast Recovery. We motivate this choice by real traffic measurements, which indicate that today’s traffic is populated by short-lived flows, whose only means to recover from a packet loss is by triggering a Retransmission Timeout. The key idea of TCP-SF can be implemented on top of any TCP lavor, from Tahoe to SACK, and requires modifications to the server TCP stack only, and can be easily coupled with recent TCP enhancements. Analytical results, simulation results, as well as testbed implementations and measurements show that TCP-SF sources outperform Classic TCP in terms of Completion Time. Also, simulations show that, when sharing a common bottleneck, TCP-SF steals little or no bandwidth from Classic TCP.
