V. Ramaswami. Traffic performance modeling for packet communication, whence where and whither. In Proc. ThirdAustralian Teletraffic Research Seminar, Melbourne, November 1988. 20  Home/Search   Document Not in Database   Summary   Related Articles   Check

....are presented by a Console. In the context of data networks measurement results of operational networks were reported in [1, 2, 3, 12] We measured performance parameters using real time traffic sources as reported in the literature in [5, 17] Theoretical work also guided our investigations [20, 9, 6]. The measurement results obtained are quantitative and reveal a great deal of information about the traffic behavior of MAGNET II. These results were qualitatively compared with measurements obtained using a class of traffic sources that exhibit both correlation between arriving packets ....

V. Ramaswami. Traffic performance modeling for packet communication whence, where, wither. Keynote address, Third Australian Teletraffic Seminar, November 1988.

....of packet switching, specifically, services from ISDN and ATM. The problem of bursty traffic packet arrival 1 modeling is among the most important challenging problems in this category. Although, numerous models of packet arrival processes have been proposed in literature, see Ramaswami et al. [9], Hellstern et al. 10] Sriram et al. 11] Heffes et al. 12] it seems that there is still a number of packet traffic features not being understood perfectly. This is partly due to uncertainties in the traffic characteristics, and to the difficulties in characterizing the traffic arrival ....

....LAN s and have observed the absence of natural length of a burst. This behavior is very different from conventional telephone traffic and from previously considered formal models of packet traffic. The same feature has also been observed in recent studies of VBR video traffic by J. Beran et al. in [9]. With the available data sets, Leland et al. investigated the persistent feature of Ethernet traffic across the network and across the time irrespective of the medium utilization level. They have graphically estimated a Hurst parameter H of about 0.80. In general, the degree of 4 self similarity ....

V.Ramaswami, "Traffic Performance Modeling for Packet Communication: Whence, Where, and Whither", Proc. of Australian Teletraffic Seminar, Nov. 1988.

....this model, the sequence generated is not Markovian. The DAR(1) model differs from the Scenic model only in that it uses an exponential scene length distribution. Markov modulated Poisson processes (MMPP) have also been proposed for modeling of various types of traffic. However, as pointed out in [12], the MMPP is not suitable for modeling traffic which tends to be periodic, as is the case for video traffic. For this reason, this model is not considered further here. Figure 1 shows the stationary distribution for the arrival process generated by the Star Wars sequence. The bar chart is the ....

V. Ramaswami. Traffic performance modeling for packet communication, whence where and whither. In Proc. Third Australian Teletraffic Research Seminar, Melbourne, November 1988.

....traffic arrival processes, and as such the problem of obtaining an accurate and tractable characterization of packet traffic is of particular importance. Packet arrival processes are so bursty and complex that we still lack a complete understanding of the features that impact performance [1]. Conventional characterizations assume that packet traffic consists of alternating active and silent periods with well defined statistics. In contrast, measurement studies have noted that there is no natural burst length, and bursts occur over many time scales [2,3] At every scale, closer ....

.... the Generalized Switched Poisson Process (GSPP) the Markov Modulated Poisson Process (MMPP) and several special cases like the Switched Poisson Process (SPP) and the interrupted Poisson process (IPP) the Markov interrupted constant rate processes, the sourcephase model and the train type model [1,4]. For reasons of mathematical tractability, most of these models retain some form of Markovian structure either in the way the arrival processes are modulated or in the arrival processes themselves. While these models have been successfully applied to the solution of many performance problems, ....

V. Ramaswami, "Traffic Performance Modeling for Packet Communication:Whence, Where and Whither," Proc. Australian Teletraffic Seminar, Nov. 1988.

....processes themselves. While these models have been successfully applied to the solution of many performance problems, they do not capture all the statistically significant features of packet traffic. One such example is the presence of serial correlations in ATM circuit emulation, and as noted in [10], such correlations can have significant impacts on performance. Another aspect of packet traffic that has so far eluded a definitive characterization is the intuitive notion of burstiness . Various statistical properties have been proposed to quantify this notion, including short term ....

V. Ramaswami, "Traffic Performance Modeling for Packet Communication:- Whence, Where and Whither," Proc. Australian Teletraffic Seminar, Nov. 1988.

....essentially the same quality as CBR coders, but at a lower average rate. In order to study the behavior of this type of source, traffic models are used. Numerous traffic models that are based on independence assumptions, e.g. Poisson streams, are not suitable for modeling video traffic (see [23, 18, 20, 16, 12]) because traffic generated by video coders is highly correlated. To overcome this problem, several models have been proposed in literature during the recent years. There are two major types of traffic model: 1. models that give a description of the sequence of the numbers of bits for each video ....

....of sources. The number of states might be increased to achieve a better fit to the correlation structure of the process. The use of the MMPP in modeling of video traffic is often criticized because it ignores the fundamental periodicity of this traffic associated with the frame rate (see e.g. [20]) A very general starting point for the modeling of video sources is presented by Latouche and Ramaswami[10] The authors do not dedicate their concept called unified stochastic model for the packet stream from periodic sources directly to the modeling of video sources, but it could be used for ....

V. Ramaswami. Traffic performance modeling for packet communication. whence, where and whither. In Proceedings of the Third Australian Teletraffic Seminar 1988, Melbourne, Nov. 1988.

No context found.

V. Ramaswami. Traffic performance modeling for packet communication, whence where and whither. In Proc. ThirdAustralian Teletraffic Research Seminar, Melbourne, November 1988. 20

No context found.

V. Ramaswami. Traffic Performance Modeling for Packet Communication Whence, Where and Whither. In Proceedings of the Third Australian Teletraffic Seminar, November 1988.

No context found.

V. Ramaswami, "Traffic Performance Modeling for Packet Communication: Whence, Where and Whither", Proc. 3rd Australian Teletraffic Seminar, Adelaide, Australia, 1988.

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