аЯрЁБс>ўџ +-ўџџџ*џџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџьЅСq` №П0 bjbjqPqP 7":: џџџџџџЄЄЄЄЄЄЄЄИ    <Иs Ж\\\\\\\\ђ є є є є є є $) h‘ – Є\\\\\ ЄЄ\\- † † † \^Є\Є\ђ † \ђ † † ЄЄ† \P @Уц Ч КF† ђ C 0s † ' p'† 'Є† l\\† \\\\\  p \\\s \\\\ИИИDќ$ИИИќИИИЄЄЄЄЄЄџџџџ An Analysis of Internet Content Delivery Systems This paper examines content delivery over Internet empirically. The paper examines content delivery from the point of view of four content delivery systems: HTTP web traffic, the Akamai content delivery network, and Kazaa and Gnutella peer-to-peer file sharing traffic. The paper uses a nine day trace with over 500 million transactions and over 20 terabytes of HTTP data in its analysis. Authors use passive network monitoring to collect traces of traffic flowing between the University of Washington (UW) and the rest of the Internet. Then they divide the data to six categories: Akamai, WWW, Gnutella, Kazaa, P2P and non-HTTP traffic. High-level characterization of the above data shows that share of HTML and GIF/JPG from HTTP traffic has decreased, while share of MP3 and AVI/MPG has increased drastically, comparing 2001 to 1999. Data also show P2P traffic has overtaken WWW traffic as the largest contributor to HTTP bytes transferred. In addition, P2P traffic makes the University a large exporter of data with video and audio accounting for a substantially larger fraction of traffic, despite the small number of requests involving those data types. Further analysis of the paper’s data shows that peer-to-peer traffic is characterized by objects whose median sizes are thousands times larger than web objects. It is this object size, rather than widespread usage, that accounts for the large fraction of bytes transferred due to P2P traffic. In fact, a small number of P2P users are consuming a high fraction of bandwidth by transferring some extremely large objects. For example, for Kazaa transfers out of the UW, the top 300 objects, with a total size of 180 GB, were responsible for 5.64 TB of the traffic – almost half of the total outbound Kazaa bandwidth. From the servers point of view, it is interesting to note that although the design of P2P overlay structures focuses on spreading the workload for scalability, a small number of servers are taking the majority of the burden according to the gathered data. Analysis also shows that both CDN caching and P2P caching can be greatly helpful. Particularly, widely deployed proxy caches would significantly reduce the need for a separate content delivery network, at least with respect to local performance. Moreover, caching would have a large effect on a wide-scale P2P system, potentially reducing wide-area bandwidth demands dramatically. In my opinion, paper’s approach to content delivery systems is very attractive. It examines real usage of different delivery systems through massive measurements and reaches to nice results specially about the fact that P2P systems can not scale well as well as role of caching in different content delivery systems. However, I believe the data of this paper should accompany data form ISP’s for commercial and residential services to be more certain about results. In general, the paper is written clear and is well-organized. It has used several graphs to explain various figures which are quite helpful. 12qruˆЈфх  ? @ B L g ˆ ‰ Ї И Й Р п р є ѕ   . / L Б _ w † ˆ ” Ё Ђ ­ Ў Ь ё љ  % 0 J K j y Џ А ѕэщлзащащащаЬШЬФНЬНЬВШЋЁЋШЋШЋШЋШ™•Ž•™Ž•Ž•‡•‡•‡•‡•‡• h`Oh`O h`OhEгh`OhЊ/ hEгhН;$hН;$6] hН;$hН;$hН;$hН;$CJaJ hЪ:hЪ:hБ/ІhН;$hЪ: hІkбhІkбh*]УhІkбhІkб6CJ]aJhІkбhІkбCJaJhhQ5CJ\aJ512A B В Г М Н $%ЂЃѓѓъъоъоеЩЩНДЈŸ“ $7$8$H$a$gdiЂ 7$8$H$gdiЂ $7$8$H$a$gdЊ/  7$8$H$gdYЂ $7$8$H$a$gdYЂ $7$8$H$a$gdТjS 7$8$H$gd`O $7$8$H$a$gd`O 7$8$H$gdН;$ $7$8$H$a$gdhQ ўА М Н    " D o p Ѕ І с ь 1^_klnpqvŽЄЅжз"#%CDJW_m}ГФщъ$…‹Œ”•ХЦ&VW“”ЂЃ'DорuљѕёъёъёъёъёъёъёъцъцтцлцлтлцлцлцлзтзцзёзёъёъёъёъётатататаЬаЬХЬХЬХНЙЕЙЕЙЕhЊ/ h ШhYЂCJaJ hYЂhYЂhYЂ hiЂhiЂhТjS h[h[hiЂh[ hXDпhXDпhXDпh`O h`Oh`OGuОПЧб ќјќє№щожвЫФНЙh{ т hІkбhІkб hЪ:hЪ: hН;$hН;$hiЂhiЂCJaJhiЂ5CJ\aJ hiЂhiЂhЊ/ hаRAhYЂh Ш  ѓъш 7$8$H$gdІkб $7$8$H$a$gdЪ:50P:phQАа/ Ар=!А"А# $ %ААаАа а†œ@@ёџ@ NormalCJ_HaJmH sH tH DA@ђџЁD Default Paragraph FontRi@ѓџГR  Table Normalі4ж l4жaі (k@єџС(No List " џџџџ12ABВГМН$%Ђ Ѓ      ˜0€€˜0€€˜0€€˜0€€˜0€€˜0€€˜0€€˜0€€˜0€€˜0€€˜0€€˜0€€˜0€€˜0€€˜0€€˜0€€˜0€€˜0€€12BВГМН$%Ђ Ѓ      Kˆ0`ЮМIˆ0Kˆ0 М(Iˆ0Iˆ0KШ0€Iˆ0 Kˆ0h!Iˆ0 Kˆ0Kˆ0Kˆ0Kˆ0€˜€€€Iˆ0€˜€€€Kˆ0 ЛА u      хы z€‘–qvљ ˜›.5CNa j 33333_wЁЂv…‹Œ”•ХЦ     љ х`O[Н;$Ъ:аRAhQТjSтTjЊ/ YЂiЂБ/ІІqЏ*]У ШІkбEгŸAиXDп{ т”Tхџ@€  ъ    @џџUnknownџџџџџџџџџџџџG‡z €џTimes New Roman5€Symbol3& ‡z €џArial"qˆ№аh“ЄЋ&VЅЋ& Зщ 1щ 1!№ ДД24d  2ƒ№HX)№џ?фџџџџџџџџџџџџџџџџџџџџџhQ2џџ0An Analysis of Internet Content Delivery SystemsAliAliўџр…ŸђљOhЋ‘+'Гй0˜дрьј , L X dpx€ˆф4An Analysis of Internet Content Delivery SystemsAliNormalAli10Microsoft Office Word@ —@ zBњ Ч@г Чщ ўџеЭеœ.“—+,љЎ0, hp˜ Ј АИРШ а фMRT www.Win2Farsi.com1 ц 1An Analysis of Internet Content Delivery Systems Title ўџџџўџџџ !ўџџџ#$%&'()ўџџџ§џџџ,ўџџџўџџџўџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџRoot Entryџџџџџџџџ РF 'Eц Ч.€1TableџџџџџџџџWordDocumentџџџџџџџџ7"SummaryInformation(џџџџDocumentSummaryInformation8џџџџџџџџџџџџ"CompObjџџџџџџџџџџџџqџџџџџџџџџџџџџџџџџџџџџџџџўџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџўџ џџџџ РFMicrosoft Office Word Document MSWordDocWord.Document.8є9Вq