SSthreshless Start: A Sender-Side TCP Intelligence for Long Fat Network
نویسندگان
چکیده
Measurement shows that 85% of TCP flows in the internet are short-lived flows that stay most of their operation in the TCP startup phase. However, many previous studies indicate that the traditional TCP Slow Start algorithm does not perform well, especially in long fat networks. Two obvious problems are known to impact the Slow Start performance, which are the blind initial setting of the Slow Start threshold and the aggressive increase of the probing rate during the startup phase regardless of the buffer sizes along the path. Current efforts focusing on tuning the Slow Start threshold and/or probing rate during the startup phase have not been considered very effective, which has prompted an investigation with a different approach. In this paper, we present a novel TCP startup method, called threshold-less slow start or SSthreshless Start, which does not need the Slow Start threshold to operate. Instead, SSthreshless Start uses the backlog status at bottleneck buffer to adaptively adjust probing rate which allows better seizing of the available bandwidth. Comparing to the traditional and other major modified startup methods, our simulation results show that SSthreshless Start achieves significant performance improvement during the startup phase. Moreover, SSthreshless Start scales well with a wide range of buffer size, propagation delay and network bandwidth. Besides, it shows excellent friendliness when operating simultaneously with the currently popular TCP NewReno connections.
منابع مشابه
TCP Congestion Window Validation
TCP’s congestion window controls the number of packets a TCP flow may have in the network at any time. However, long periods when the sender is idle or application-limited can lead to the invalidation of the congestion window, in that the congestion window no longer reflects current information about the state of the network. In this paper we propose a simple modification to TCP’s congestion co...
متن کاملA Simple Refinement of Slow-Start of TCP Congestion Control
This paper presents a new variant of Slow-start, called Smooth-start, which provides a smooth transition between the exponential and linear growth phases of TCP congestion window. Slow-start is known to make an abrupt transition between the Slow-start and CongestionAvoidance phases, and hence, often causes multiple packet losses from a window of data and retransmission timeouts, which, in turn,...
متن کاملTCP Westwood: congestion control with faster recovery
In this paper we propose a new version of the TCP protocol, called TCP Westwood, which enhances the performance of TCP window congestion control by using as feedback the end-to-end measurement of the bandwidth available along a TCP connection. The available bandwidth is estimated at the TCP source by measuring and low-pass filtering the returning rate of acknowledgments. The estimated bandwidth...
متن کاملA New Scheme for TCP Congestion Control: Smooth-Start and Dynamic Recovery
This paper presents a new approach to TCP congestion control. The new scheme includes two parts: (1) the Smooth-Start algorithm, which replaces the Slow-Start algorithm at the start of a TCP connection or after a retransmission timeout, and (2) the Dynamic Recovery algorithm, which replaces the Fast Recovery algorithm to recover packet losses when a TCP connection is congested. Both algorithms ...
متن کاملHybrid Slow Start for High-Bandwidth and Long-Distance Networks
Slow Start is a technique to probe for unknown and time-varying available bandwidth of a network path. A sender increases its congestion window by one for each ACK received (when ACKs are not delayed), which effectively doubles its congestion window when receiving ACKs for all the packets within a congestion window. Even if an exponential increase of congestion window during Slow Start grabs un...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
- CoRR
دوره abs/1401.7146 شماره
صفحات -
تاریخ انتشار 2010