ࡱ> -/,q`bjbjqPqP 2::XXXlt t t t lR( $zhX    fffL8 X fff5@Xu ЏsVt ()"0R=8bFb(ubXu@ Y |f d9 V RlllD lll lll Statistical properties of PMA traffic traces and application to capacity sizing Zdenko Vrdoljak University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb zdenko.vrdoljak@fer.hr Marije Ljolje University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb marije.ljolje@fer.hr Abstract: In this paper we describe results from the joined project with the aim to analyze statistical properties of publicly available IP traffic traces and study the influence of the properties to traffic sizing procedure. The analysis is based on the network traces database provided by the Passive Measurement and Analysis (PMA) project of the National Laboratory for Applied Network Research (NLANR) of USA. We establish that the traffic rate distribution can be closely described by the sub-exponential distributions and that the time-dependence is strong. The most important result is that the packet delays observed on different nodes in the network belonging to the same traffic flow can be analyzed as (highly) independent random variables. This is especially the truth when the capacity of the links in the network are chosen in such a manner that maximum 1% of all packets have delay higher than 10 ms. This conclusion motivates us to build a fully empiric capacity calculation formula that can be applied on all links in the network knowing the traffic matrix. The formula takes into account only the mean value of the traffic rate what makes it very simple to use. Keywords: Internet traffic, modeling, capacity sizing PQRa1 2 5 : D ] ^ , 5 8 a ÿǟh$$@ h5h?h0hs26 h0hs2h4hs2hT|hj@ h56h?hmH sH h?h$$@mH sH h$$@mH sH hhmH sH hj@mH sH -QRbx$ 9 : gd$$@0&P P. A!"#$% H@H Normal$7$8$a$CJ_HmH sH tH P@P Heading 1$$@&a$5CJ\aJmH sH J@J Heading 2$@&6CJ]aJmH sH r@r Heading 30$$$d&d@&NPa$6CJ]aJmH sH :@: Heading 4$@&5\DAD Default Paragraph FontViV  Table Normal :V 44 la (k(No List <O< References  & F aJHOH TitleofPaper$a$5CJ\aJ0O0 NormalTextDO"D SectionTitles$a$mH sH JO!2J SubsectionTitles$a$mH sH HOBH AuthorsNames$a$5CJ\aJLORL AuthorsAffiliations$a$aJFObF AuthorsE-mail$a$ 6]aJPO1rP AbstractAndKeywords\]mH sH HOH TitlesOfTables$a$ 5\aJH@H ? Balloon TextCJOJQJ^JaJ QRbx$9:000000000000(0(0(0000$9:K00ȣI00I000I00I00   l,2$&%􊉢hzb4wcy@0(  B S  ? + T.5 , \- \. /