727 KB24 KB198520252008Čučej, ŽarkoFras, MatjažMohorko, Jožeštevilka:2letnik:38str. 117-123ISSN:0352-9045COBISSID:12946710URN:URN:NBN:SI:doc-IURA1EMGslStrokovno društvo za mikroelektroniko, elektronske sestavne dele in materialeInformacije MIDEMmodeliranjeomrežni prometsimulacijetelekomunikacijska omrežjavrednotenjezmogljivostAnaliza, modeliranje in simulacija vpliva prometa aplikacij za izmenjavo datotek P2P na zmogljivost omrežij| Analysis, modeling and simulation of P2P file sharing traffic impact on networks' performances|Last year's CachneLogic research shows that, in January 2006, P2P traffic accounted for approximately 71% of all internet traffic, and was the main reason for internet traffic increasing. Figure 1 shows the trend of P2P traffic increase. The pioneer of P2P file sharing applications is very well-known Napster, which was created in 1999 by Shawn Fanning and used for sharing music files. Napster was finally destroyed by the music industry after Court proceedings. Napster, as a concept, had a great influence on developing file-sharing applications, such as eMula, Kazza, pTorrent, UmeWire, Skype etc. In our research, we paid attention to the modeling of network traffic caused by P2P file sharing applications (P2P traffic). We measured P2P traffic using the Wireshark network traffic capture tool as shown in Figure 2. We created a self-similarity and long-range dependence analysis for the measured tests P2P traffics, the results of which are shown in Table 1, and in Figures 3 and 4. We also estimated the parameters for a statistical description of a P2P network separately for the processes of packet-size and inter-arrival time, which represent the main network traffic processes. The histogram method for distribution parameters' estimation is shown in Figure 5. These estimated distribution parameters were used for modeling a customized application of P2P network traffic in OPNET simulation tool. Using simulations with simple test networks, as shown in Figure 6, we represented the possibilities of simulating the impact of P2P network traffic on network performances. We show how the volume of P2P users impacts on other applications' performances (web time response shown in Figure 7 and link utilization shown in Figure 8.)V zadnjih letih doživlja delež prometa, ki ga ustvarjajo P2P aplikacije za izmenjavo datotek skokovito rast. S svojo količino in intenzivnostjo znatno vpliva na zmogljivost omrežij. Namen članka je predstavitev metode za analizo in modeliranje takšnega prometa za potrebe simulacij. Na osnovi izmerjenega prometa smo ocenili statistične parametre za modeliranje tovrstnega prometa. V simulacijskem okolju OPNET smo zgradili testno omrežje, kjer se pretakajo poleg ostalih prometov še promet aplikacij P2P, ki smo ga modelirali s predlagano metodo. S simulacijami smo preučili vpliv P2P prometa na zmogljivost omrežjaTEXTznanstveno časopisjejournalsSlovenian National E-content AggregatorNational and University Library of SloveniaStrokovno društvo za mikroelektroniko, elektronske sestavne dele in materiale