10 citations found. Retrieving documents...
I. Cidon, R. Guerin, and A. Khamisy. Protective Buffer Management Policies. IEEE/ACM Transactions on Networking, 2(3), June 1994.

 Home/Search   Document Not in Database   Summary   Related Articles   Check  

This paper is cited in the following contexts:
BLUE: A New Class of Active Queue Management Algorithms - Feng, Kandlur, Saha, Shin (1999)   (79 citations)  (Correct)

....control can be performed with a minimal amount of buffer space. This reduces the end to end delay over the network, which in turn, improves the effectiveness of the congestion control algorithm. In addition, smaller buffering requirements allow more memory to be allocated to high priority packets [4, 15], and frees up memory for other router functions such as storing large routing tables. Finally, BLUE allows legacy routers to perform well even with limited memory resources. 3.3 Understanding BLUE To fully understand the difference between the RED and BLUE algorithms, Figure 6 compares their ....

I. Cidon, R. Guerin, and A. Khamisy. Protective Buffer Management Policies. IEEE/ACM Transactions on Networking, 2(3), June 1994.


Building Predictable Services Using AF - Feng   (Correct)

....the AF PHB becomes meaningless when priority packets are not given sufficient protection from non priority packets. In order to address this problem, a number of queue management algorithms have been proposed which advocate the strict allocation of buffer resources to packets marked with AF [10, 11]. The idea behind these algorithms is to control the sharing of buffer resources between priority and non priority packets in order to minimize the number of priority packets which are lost. 4 packet One priority packet One priority packet One priority packet IN OUT IN OUT non priority ....

I. Cidon, R. Guerin, and A. Khamisy, "Protective Buffer Management Policies," IEEE/ACM Transactions on Networking, vol. 2, no. 3, , June 1994.


BLUE: An Alternative Approach to Active Queue Management - Feng, Kandlur, Saha, Shin   (6 citations)  (Correct)

....utilization. The most important consequence of using BLUE is that congestion control can be performed with a minimal amount of buffer space. thus reducing the endto end delay over the network. In addition, smaller buffering requirements allow more memory to be allocated to high priority packets [3], and frees up memory for other router functions such as storing large routing tables. Finally, BLUE allows legacy routers to perform well even with limited memory resources. C. Understanding BLUE To fully understand the difference between the RED and BLUE algorithms, Figure 6 compares their ....

I. Cidon, R. Guerin, and A. Khamisy. Protective Buffer Management Policies. IEEE/ACM Transactions on Networking, 2(3), June 1994.


The BLUE Active Queue Management Algorithms - Feng, Shin, Kandlur, Saha (2002)   (5 citations)  (Correct)

....control can be performed with a minimal amount of buffer space. This reduces the end to end delay over the network, which in turn, improves the effectiveness of the congestion control algorithm. In addition, smaller buffering requirements allow more memory to be allocated to high priority packets [5, 16], and frees up memory for other router functions such as storing large routing tables. Finally, BLUE allows legacy routers to per 0.0 50.0 100.0 150.0 200.0 Buffer Size (in ms of delay) 0.0 5.0 10.0 15.0 B2 B3 B4 R1 R2 R3 R4 0.0 50.0 100.0 150.0 200.0 Buffer Size (in ms of delay) ....

I. Cidon, R. Guerin, and A. Khamisy. Protective Buffer Management Policies. IEEE/ACM Transactions on Networking, 2(3), June 1994.


Improving Internet Congestion Control And Queue Management.. - Feng (1999)   (11 citations)  (Correct)

....control can be performed with a minimal amount of buffer space. This reduces the end to end delay over the network, which in turn, improves the effectiveness of the congestion control algorithm. In addition, smaller buffering requirements allow more memory to be allocated to high priority packets [10, 30], and frees up memory for other router functions such as storing large routing tables. Finally, BLUE allows legacy routers to perform well even with limited memory resources. 4.3.3 Understanding BLUE To fully understand the difference between the RED and BLUE algorithms, Figure 4.6 compares ....

....such as max th are triggered by an average queue length calculation, these settings can still lead to unnecessary losses in compliant packets. For example, additional queueing work has shown the use of absolute thresholds for limiting unmarked packets can effectively prevent loss of marked packets [10, 30]. While ERED works well for the experiments in this study, a queue management algorithm which reserves a portion of the queue for marked packets while performing RED on the remaining portion might be ideal for controlling congestion while effectively supporting the priority marking. An ....

I. Cidon, R. Guerin, and A. Khamisy. Protective Buffer Management Policies. IEEE/ACM Transactions on Networking, 2(3), June 1994.


BLUE: A New Class of Active Queue Management Algorithms - Dilip, Kandlur, Saha, Shin (1999)   (3 citations)  (Correct)

....control can be performed with a minimal amount of buffer space. This reduces the end to end delay over the network, which in turn, improves the effectiveness of the congestion control algorithm. In addition, smaller buffering requirements allow more memory to be allocated to high priority packets [4, 15], and frees up memory for other router functions such as storing large routing tables. Finally, BLUE allows legacy routers to perform well even with limited memory resources. 3.3 Understanding BLUE To fully understand the difference between the RED and BLUE algorithms, Figure 6 compares their ....

I. Cidon, R. Guerin, and A. Khamisy. Protective Buffer Management Policies. IEEE/ACMTransactions on Networking, 2(3), June 1994.


Quality-of-Service in Packet Networks Basic Mechanisms and.. - Guérin, Peris (1999)   (37 citations)  (Correct)

....in, and as a result some packets from an arriving high priority burst are lost. One solution is to allow the high priority packets to push out [42] the low priority ones, but this capability adds to the implementation complexity of the scheme. In addition, it may not always be sufficient (see [7] for a discussion of this issue) and additional mechanisms may be needed to try to avoid congestion. In this section, we describe a few such buffer management mechanisms like Early Packet Discard (EPD) 51, 59] and Random Early Discard (RED) 18] where, as a preventive measure, packets are ....

I. Cidon, R. Gu'erin, and A. Khamisy. Protective buffer management policies. IEEE/ACM Trans. Networking, 2(3):240--246, June 1994.


Understanding and Improving TCP Performance over.. - Feng, Kandlur, Saha.. (1997)   (18 citations)  (Correct)

....occupancy and max th (unmark) L, the time needed to drain the unmarked packets. Unfortunately, since the thresholds such as maxth are triggered by an average queue length calculation, these settings can still lead to unnecessary losses in compliant packets. For example, additional queueing work [2] has shown the use of absolute thresholds for limiting unmarked packets can effectively prevent loss of marked packets. While ered works well for the experiments in this study, a queue management algorithm which reserves a portion of the queue for marked packets while performing red on the ....

I. Cidon, R. Guerin, and A. Khamisy. Protective Buffer Management Policies. IEEE/ACM Transactions on Networking, 2(3), June 1994.


Scalable QoS Provision Through Buffer Management - Guerin Kamat Peris (1998)   (32 citations)  Self-citation (Guerin)   (Correct)

....usually consists of some global state information, e.g. the total buffer content, as well as additional state information specific to the flow to which the packet belongs, e.g. the number of packets the flow currently has in the buffer. For example, this is true for threshold based mechanisms [2], schemes such as Early Packet Discard (EPD) 7, 9] Random Early Discard (RED) 3] and Fair RED (FRED) 5] Scheduling decisions, on the other hand, require both flow specific state information, e.g. the last transmission time of a packet from the flow, and operations involving all the other ....

I. Cidon, R. Guerin, and A. Khamisy. Protective buffer management policies. IEEE/ACM Trans. Networking, 2(3):240-- 246, June 1994.


The plaNET/ORBIT High Speed Network - Cidon, Gopal, Gopal, Guerin.. (1992)   (6 citations)  Self-citation (Cidon Gurin)   (Correct)

....the regular traffic. Specifically, excess traffic is sent at low loss priority and is discarded whenever the backlog in a network queue exceeds a given threshold. This guarantees a certain level of protection to regular traffic, irrespective of the intensity or arrival patterns of excess traffic [26]. 4.0 THE BACKBONE SWITCH The wide area backbone network is composed of interconnected plaNET switches as depicted in Figure 1. A key aspect of plaNET is that all intermediate packet handling functions are performed on the fly in hardware without software involvement. This enables operation at ....

I. Cidon, R. Gurin and A. Khamisy, "Protective Buffer Management Policies," Proc. INFOCOM'93, pp. 1051-1058, San Francisco, CA, March 1993.

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