Robert Morris. Scalable TCP congestion control. In Proceedings of IEEE INFOCOM 2000. Home/Search Document Details and Download
Summary Related Articles Check |
This paper is cited in the following contexts:
First 50 documents
An Active Queue Management Scheme Based on a Capture-Recapture.. - Chan, Hamdi (2003) (Correct)
....space for the other flows. In this case, can be calculated as follows: ## is the buffer size and is the total number of flows in the buffer. As is a predefined value, estimating the appropriate value requires the estimation of the number of flows ( buffer. Moreover, in [1] and [2], the estimation of the number of flows has been shown to be an important information for an AQM mechanism. Based on the ### CR model, we can estimate by considering the total number of the flows in the buffer as the total of the animals in the population. For example, there are flows in the ....
R. Morris, "Scalable TCP Congestion Control," Ph.D. Thesis, Harvard University, 1998. 111 pages.
Queue Management for Short-Lived TCP Flows in Backbone Routers - Kantawala, Turner (2002) (Correct)
....if each TCP flow is able to obtain its share of the link bandwidth, the end to end delay remains very high. This is exacerbated for flows with multiple hops, since packets may experience high queueing delays at each hop. This phenomenon is well known and has been discussed by Hashem [1] and Morris [2], among others. To address this issue, researchers have developed alternative queueing algorithms which try to keep average queue sizes low, while still providing high throughput and link utilization. The most popular of these is Random Early Discard or RED [3] RED maintains an ....
Robert Morris, "Scalable TCP Congestion Control", in IEEE INFOCOM 2000.
Generating Prescribed Traffic with HTTP/TCP Sources for Large.. - Below, Killat (2003) (1 citation) (Correct)
....matches the prescribed (i.e. measured) traffic by the network provider. The number of sources for each node and flow has to be determined in order to solve this problem, which produce the known, observed traffic. This is not a trivial problem since the steady state formulas for the TCP throughput [4] are not applicable for short lived connections that are characteristic for WWW traffic. We investigate this problem for HTTP applications and thus face the problem of the reactive behavior of TCP. The intrinsic features of the WWW traffic within the last years are the user behavior and the use ....
Robert Morris, "Scalable TCP congestion control," in IEEE INFOCOM 2000, Tel Aviv, Mar. 2000, pp. 1176--1183.
Efficient Queue Management for TCP Flows - Kantawala, Turner (2001) (Correct)
....if each TCP flow is able to obtain its share of the link bandwidth, the end to end delay remains very high. This is exacerbated for flows with multiple hops, since packets may experience high queueing delays at each hop. This phenomenon is well known and has been discussed by Hashem [1] and Morris [2], among others. To address this issue, researchers have developed alternative queueing algorithms which try to keep average queue sizes low, while still providing high throughput and link utilization. The most popular of these is Random Early Discard or RED [3] RED maintains an ....
Robert Morris, "Scalable TCP Congestion Control", in IEEE INFOCOM 2000.
Design of a High Performance Dynamically Extensible.. - Kuhns, DeHart.. (2002) (1 citation) (Correct)
....round trip delays. Such buffers can delay packets for as much as half a second during congestion periods. When such large queues carry heavy TCP traffic loads and are serviced with the Tail Drop policy, the large queues remain close to full resulting in high end to end delays for long periods [30, 31]. RED maintains an exponentially weighted moving average of the queue length, an indicator of congestion. When the average crosses a minimum threshold ( ##### # ) packets are randomly dropped or marked with an explicit congestion notification (ECN) bit. When the queue length exceeds the ....
R. Morris, "Scalable TCP Congestion Control," in IEEE INFOCOM 2000.
Design and Evaluation of a High-Performance.. - Kuhns, DeHart.. (2002) (Correct)
....round trip delays. Such buffers can delay packets for as much as half a second during congestion periods. When such large queues carry heavy TCP traffic loads and are serviced with the Tail Drop policy, the large queues remain close to full resulting in high end to end delays for long periods [33, 34]. RED maintains an exponentially weighted moving average of the queue length and uses this as an indicator of congestion. When the average crosses a minimum threshold ( # ### ) packets are randomly dropped or marked with an explicit congestion notification (ECN) bit. When the queue length ....
R. Morris, "Scalable TCP Congestion Control," in IEEE INFOCOM 2000.
On the Configuration of Simulations of Large Network Models.. - Below, Killat (2002) (Correct)
.... selection of the distributions and their parameter fittings for the generation of self similar traffic can be found in [8, 3, 9] One major concern for the simulation of large networks is the establishment of the required traffic intensities since the steady state formulas for the TCP throughput [10] are not applicable for short time connections that are characteristic for WWW traffic. Moreover, even though computing power has increased considerably, simulation of the real number of clients with a large network model seems to be impossible, at least if a large number of simulations needs to ....
....propose an algorithm to distribute these sources among all flows in order to produce traffic according to the traffic matrix TP as follows. Although we know that TCP is not in steady state we use the proportional relation r src (RTT # p loss ) 1 as an approximation for the average source rate [10]. With the approximation of a homogeneous loss probability for all flows we get RTT = const. 12) We use Equation (12) to allocate the number of sources proportional to the throughput RTT product: we build a normalized traffic matrix A and a normalized flow RTT matrix D a i j = t p i j t ....
Robert Morris, "Scalable TCP congestion control, " in IEEE INFOCOM 2000.
Queue Management for Short-Lived TCP Flows in Backbone Routers - Kantawala, Turner (2000) (Correct)
....if each TCP flow is able to obtain its share of the link bandwidth, the end to end delay remains very high. This is exacerbated for flows with multiple hops, since packets may experience high queueing delays at each hop. This phenomenon is well known and has been discussed by Hashem [1] and Morris [2], among others. To address this issue, researchers have developed alternative queueing algorithms which try to keep average queue sizes low, while still providing high throughput and link utilization. The most popular of these is Random Early Discard or RED [3] RED maintains an ....
Robert Morris, "Scalable TCP Congestion Control", in IEEE INFOCOM 2000.
Extended Analysis of Binary Adjustment Algorithms - Gorinsky, Vin (2002) (6 citations) (Correct)
....shows for Chiu Jain model, oscillations of the total load after convergence under AIMD grow linearly in size as the number of users increases. Morris confirms experimentally that the average loss rate grows linearly with the number of competing TCP connections and thereby worsens TCP performance [11]. On the other hand, MIMD congestion control is scalable since the size of the total load oscillations after convergence under MIMD does not depend on the number of users. Consequently, it is possible to design such an MIMD controlled network with ECN style packet marking [14] that traffic in ....
R. Morris. Scalable TCP Congestion Control. In Proceedings IEEE INFOCOM 2000.
Internet Traffic Engineering - Mortier (2002) (2 citations) (Correct)
....or transmission bandwidth. If it is not detected and prevented, then congestion collapse may occur; this is where the network, or some subset of the network, is loaded to such a level that goodput the throughput of data, disregarding retransmissions falls to negligible levels [Jacobson88, Morris99] Following the rapid increase in the use of TCP and enormous changes in the topology and size of the Internet, a succession of congestion control mechanisms have been proposed and implemented for TCP. These began with the Slow Start and Congestion Avoidance schemes [Jacobson88 ] and ....
....more smoothly, i.e. at a finer granularity. 3.1. 1 TCP congestion collapse Although these methods give satisfactory performance in many cases, it is still the case that a TCP flow may observe near zero goodput when a large number of TCP flows share a bottleneck link in the Internet [Morris97, Morris99] The consequent competition for resources results in catastrophic collapse of the per flow performance, even though the link is operating at full utilization. Congestion collapse occurs as each TCP flow probes for available bandwidth to see if it may increase the amount of data it has ....
R. Morris. Scalable TCP Congestion Control. PhD thesis, Harvard University, January 1999. (pp 15, 40)
Measurement-Based Management of Network Resources - Moore (2002) (Correct)
....first drawback of the static allocation approach towards TCP would be that changes in the number of flows would not be reflected in changes in the allocation. A solution may be to use the number of flows present to compute an appropriate allocation. A simple mechanism, such as that proposed in [Morris00] caches TCP header information on source and destination to compute an approximation of the number of active flows. Measurements of utilisation of a limited number of TCP IP flows will produce informative results on the activity of those sessions, such as the activity cycle a single web browser ....
Robert Morris. Scalable TCP Congestion Control. In Proceedings of IEEE INFOCOM 2000, Tel Aviv, Israel, March 2000. (p 234) 265
Recent Results on Sizing Router Buffers - Appenzeller, McKeown, Sommers.. (Correct)
No context found.
Robert Morris. Scalable TCP congestion control. In Proceedings of IEEE INFOCOM 2000.
Differentiated Predictive Fair Service for TCP Flows - Matta, Guo (2000) (3 citations) (Correct)
No context found.
R.T. Morris, "Scalable TCP Congestion Control," in IEEE/INFOCOM'2000.
On the Advantages of Lifetime and RTT Classification Schemes.. - Wu, Nikolaidis (2002) (Correct)
No context found.
R. Morris, "Scalable TCP Congestion Control," in Proc. of INFOCOM 2000.
Sizing Router Buffers - Appenzeller, Keslassy, McKeown (2004) (5 citations) (Correct)
No context found.
R. Morris. Scalable tcp congestion control. In Proceedings of IEEE INFOCOM 2000.
Tradeoffs of Implementing TCP Fairness Schemes - Wu, Nikolaidis (2003) (Correct)
No context found.
R. Morris, "Scalable TCP Congestion Control, " in Proc. INFOCOM 2000.
Dynamic Bandwidth Allocation for Lifetime-Based TCP.. - Wu, Nikolaidis (2003) (Correct)
No context found.
R. Morris, "Scalable TCP Congestion Control," in Proc. of INFOCOM 2000.
Analytic Models for the Latency and Steady-State.. - Sikdar, Kalyanaraman, .. (2001) (3 citations) (Correct)
No context found.
R. Morris, "Scalable TCP congestion control," Proceedings of IEEE INFOCOM, pp. 1176-1183, Tel-Aviv, Israel, March 2000.
Extended Analysis of Binary Adjustment Algorithms - Sergey Gorinsky Harrick (2002) (6 citations) (Correct)
No context found.
R. Morris. Scalable TCP Congestion Control. In Proceedings IEEE INFOCOM 2000.
Sizing Router Buffers - Appenzeller, Keslassy, McKeown (2004) (5 citations) (Correct)
No context found.
R. Morris. Scalable tcp congestion control. In Proceedings of IEEE INFOCOM 2000.
On Achieving Weighted Service Differentiation: An.. - Hsieh, Kim, Sivakumar (2003) (Correct)
No context found.
Morris, R.: Scalable TCP congestion control. In: Proceedings of IEEE INFOCOM, Tel-Aviv, Israel (2000)
Sizing Router Buffers - Appenzeller, Keslassy, McKeown (2004) (5 citations) (Correct)
No context found.
R. Morris. Scalable tcp congestion control. In Proceedings of IEEE INFOCOM 2000.
TCP with Sender-based Delay Control - Kung Koan-Sin Tan (2003) (Correct)
No context found.
R. T. Morris, Scalable TCP congestion control, Ph.D. thesis, Harvard University (1999).
Effairness: A Metric for Congestion Control Evaluation in .. - Sergey Gorinsky Harrick (2001) (Correct)
No context found.
R. Morris. Scalable TCP Congestion Control. In Proceedings IEEE INFOCOM 2000.
The Transmission Control Protocol - Noureddine, Tobagi (2002) (Correct)
No context found.
Morris R., Scalable TCP Congestion Control, in Proceedings of INFOCOM, March 2000.
First 50 documents
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC