J. B. Kim, R. Simha and T. Suda, "Analysis of a Finite Buffer Queue with Heterogeneous Markov Modulated Arrival processes: A Study of Traffic Burstiness and Priority packet Discarding," Computer Networks and ISDN Systems 28 pp. 653-673, 1996. Home/Search Document Details and Download
Summary Related Articles Check |
This paper is cited in the following contexts:
Performance Modeling and Management of High-Speed Networks - Gupta (1993) (24 citations) (Correct)
....cell loss probability acceptable to different classes is not the same we shall see in Chapter 3 that it is a reasonable assumption. 11 In general the cell loss probability for various classes of VCs is different. It has been shown that the cell loss probability is proportional to the burstiness [5]. CHAPTER 2. ROUTING IN ATM NETWORKS 58 assigned to route fjg or to route fi; jg, respectively. Consider setting up a directly routed VC b j . The QOS requirement will permit this if the following two conditions are satisfied: D1) p j (B j [ fbg) ffl ; D2) For every VP k such that there is ....
....the demands of all VCs. For simplicity consider the case of two source types with B 1 B 2 . It is well known that the cell loss probability for Source 2 (with less burstiness) is less than the overall cell loss probability and that for Source 1 is more than the overall cell loss probability [5]. Further, the difference between the loss probabilities of the two sources increases with the difference in their burstiness. Let (D; ffl) be the QOS that needs to be ensured so that the QOS demands of both sources are met. The delay bound, D, is chosen as the minimum of the individual delay ....
[Article contains additional citation context not shown here]
J. J. Bae, T. Suda, and R. Simha. Analysis of a finite buffer queue with heterogeneous markov modulated arrival process: A study of the effects of traffic burstiness on individual packet loss. In Proceedings of the IEEE INFOCOM, pages 219--230, 1992.
An Approach towards End-to-end QoS with Statistical.. - Rampal, Reeves, Agrawal (1995) (1 citation) (Correct)
....in their allowable cell loss (more than 4 orders of magnitude) Bandwidth partitioning is also more efficient when a class with more stringent cell loss constraint but lower arrival rate is multiplexed with a traffic class which can tolerate higher cell loss but has higher arrival rate. In [2], Bae et al. show that when multiplexing a heterogeneous set of sources, the allowable CLP may have to be set more stringent than the most stringent of the individual allowable cell losses in order to meet the QoS requirements of Q3 Q2 Q1 T1=10 T2=14 T3=43 Arrivals at Q1 Departures from Q1 ....
J.J. Bae, T. Suda and R. Simha, "Analysis of a Finite Buffer Queue with Heterogeneous Markov Modulated Arrival Processes: A Study of the Effects of Traffic Burstiness on Individual Packet Loss," in Proc. IEEE INFOCOM '92, 1992, pp.219-230.
Loss Correlation for Queues with Bursty Input Streams - Schulzrinne, Kurose, Towsley (1992) (17 citations) (Correct)
....of the sources with Bernoulli arrivals, at least for large N . These assumptions remain to be validated by analysis. Also, the cases of several heterogeneous sources and of sources with two different Bernoulli arrival processes (MMBP) rather than on off sources deserve closer scrutiny (see [24, 25] for a loss probability analysis) In the Gilbert channel model [1] the system alternates between two states in a manner identical to the IPP source. In one state, every packet is lost, in the other, no losses occur. While the loss runs are indeed geometrically distributed, simulations indicate ....
J. J. Bae, T. Suda, and R. Simha, "Analysis of a finite buffer queue with heterogeneous markov modulated arrival processes: A study of the effect of traffic burstiness on individual packet loss." submitted to Infocom 1992, May 1992.
End-to-end guaranteed QoS with statistical multiplexing.. - Rampal, Reeves, Agrawal (1 citation) (Correct)
....to be work conserving. We would like to statistically multiplex all the different traffic types together to obtain the maximummultiplexing gain. However, several studies have shown that different traffic types which vary widely in burstiness do not multiplex very well. As an example, Bae et al. [1] have shown that the QoS specification for a set of heterogeneous sources may have to be set more stringent than the most stringent of any of the individual QoS requirements to ensure all sources receive their required QoS. Hence a mechanism for separating out different traffic types seems ....
J.J. Bae, T. Suda and R. Simha, "Analysis of a Finite Buffer Queue with Heterogeneous Markov Modulated Arrival Processes: A Study of the Effects of Traffic Burstiness on Individual Packet Loss," in Proc. IEEE INFOCOM '92, pp.219-230, 1992.
Routing And End-To-End Quality Of Service In Multimedia Networks - Rampal (1995) (8 citations) (Correct)
....either bandwidth sharing or bandwidth partitioning can lead to lower total bandwidth requirement for a set of sources depending on different scenarios of relative burstiness, QoS requirements, arrival rates of the sources being multiplexed and the cell scheduling policy used. See for example [10] [4], 6] Additionally, the gains achieved by statistical bandwidth sharing typically diminish as the multiplexing level is increased. For instance Figure 3.6 shows the total bandwidth requirement for a set of 200 voice sources at different partitioning levels; an approximation based on Gaussian ....
J.J. Bae, T. Suda and R. Simha, "Analysis of a Finite Buffer Queue with Heterogeneous Markov Modulated Arrival Processes: A Study of the Effects of Traffic Burstiness on Individual Packet Loss," in Proc. IEEE INFOCOM '92, 1992, pp. 219-230.
Performance Analysis of Cell Switching Management.. - Communications Jongho.. (Correct)
No context found.
J. B. Kim, R. Simha and T. Suda, "Analysis of a Finite Buffer Queue with Heterogeneous Markov Modulated Arrival processes: A Study of Traffic Burstiness and Priority packet Discarding," Computer Networks and ISDN Systems 28 pp. 653-673, 1996.
Performance Analysis of an ATM MUX with a New Space.. - Bang, Ansari, Tekinay (2002) (Correct)
No context found.
J. Kim, R. Simha, and T. Suda, "Analysis of a finite buffer queue with heterogeneous markov modulated arrival processes: A study of traffic burstiness and priority packet discarding," Computer Networks and ISDN Systems 28, pp. 653--673, 1996.
Per-flow Delay Performance in Traffic Aggregates - Siripongwutikorn, Banerjee (2002) (Correct)
No context found.
J.J. Bae, T. Suda, and R. Simha, "Analysis of a Finite Buffer Queue with Heterogeneous Markov Modulated Arrival Process: A Study of the Effects of Traffic Burstiness on Individual Packet Loss," in Proc. IEEE INFOCOM'92, Florence, Italy, 1992, pp. 219--230.
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