This sub section shows the traffic distribution of the residential broadband applications’ quality scenario. Figure 4.2 below represents the contribution of the traffic generated by each application type to the overall network’s traffic. It is clear from this figure that P2P file sharing application generates more than half of the total network’s traffic. Web browsing on the other hand, represents the second most throughput generating application. The dominance of P2P file sharing traffic is attributed to the high per user bit rate consumption of such applications (see Table 4.2), and the large number of entities sharing the files, where if a particular user demands a certain file, the file should be assembled from all the other users this file is partitioned and distributed among them. Consequently, extra traffic will be added to the network.
Figure 4.2 Total Network's Load Rate due to the Different Application Types
On the other hand, web browsing occupies the second rank in terms of bandwidth consumption because of the popularity of such applications, where about 75% users use this application at any particular time. This reason together with the high per user throughput generated by this applications type (see Table 4.2) makes web browsing the second dominant application in this scenario.
Figure 4.3 above represents the total throughputs on the various network’s links at busy hour. It is clear that links [Basra-Ashar] and [Server-Basra] are the most congested links. The first link is congested because the source node (Basra) and the destination node (Ashar) have the highest populations among the network’s nodes. The second link on the other hand (Server-Basra) is congested because this link serves as the only gateway for the bulky web browsing traffic to get downloaded to all the other network’s nodes.
Figure 4.4 shows that P2P file sharing throughput at busy hour dominates the traffic over the link [Ashar-Basra] to the extent that the traffic generated by the other applications can be neglected compared to this application’s traffic. A question may arise here, where the traffic of the competitive application (web browsing) has gone? The answer is obvious in Table 4.2 and the topology of the network shown in Figure 4.1 where this link holds only the upload component of the web browsing traffic to the server which has very a low per user busy hour bit rate as shown in Table 4.2. Note that no loads are transmitting on this link due to streaming audio, streaming video, and IPTV applications since these applications generate one way traffics from the server to the other nodes (opposite to this link’s direction).
Figure 4.4 Busy Hour Load Rates on the Link Ashar-Basra due to the Different Application Types
The hourly distributions of flow due to the different application types together with the resulting total per hour flow (coloured red) are shown in Figure 4.5. It is obvious in this figure that the per-hour total rate on the link [Ashar-Basra] largely inherits the shape of P2P traffic’s distribution since it is the only dominant application on this link.
Figure 4.5 Per Hour Load Rate on Link Ashar-Basra
The busy hour bandwidth consumption contribution of the different applications’ traffics on the two links [Basra-Ashar] and [Basra-Zubair] will include the bulky bandwidth consuming component of web browsing (download) causing the effect of the competitive application (web browsing) to reappear on the overall link’s loading as shown in Figures 4.6 and 4.7 respectively.
Figure 4.6 Busy Hour Load Rates on the Link Basra-Ashar due to the Different Application Types
Although the bandwidth consumed by the web browsing applications is substantial here on the links [Basra-Ashar] and [Basra-Zubair], the per hour total load rate distributions on these links shown in Figures 4.8 and 4.9 respectively is still dominated by the P2P file sharing since it generates much more throughput than the web browsing.
Figure 4.8 Busy Hour Load Rates on the Link Basra-AlZubair due to the Different Application Types
Figure 4.9 Per Hour Load Rate on Link Basra-Al-Zubair
The loading pattern on the link [Server-Basra] due to the various applications types shown in Figure 4.10 is different from the loading patterns of the other links since only four applications contribute to construct the download of the link. These four applications are those they send broadcasting data from the server to other nodes, i.e., web browsing streaming audio streaming video, and IPTV.
Other applications are characterised to be point to point based where they do not need to have their traffic sent/received to/from the server. In other words, they do not need server contribution in their communications. Therefore, they do not send traffic through this link.
Figure 4.10 Busy Load Rates on the Link Server-Basra due to the Different Application Types
Since no P2P traffic exists on the link [Server-Basra], web browsing applications will play the role of the dominant applications type in terms of bandwidth consumption. As a result, the per hour total flow distribution will take the shape of web browsing per hour load distribution as shown in Figure 4.11.