Jeff Sedayao. "Mosaic Will Kill My Network!" -- Studying Network Traffic Patterns of Mosaic Use. In Electronic Proceedings of the Second World Wide Web Conference '94: Mosaic and the Web, Chicago, Illinois, October 1994.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....lines, while the lognormal dataset is characterized by lines that seem roughly convergent. 3 Related Work The first step in understanding WWW traffic is the collection of trace data. Previous measurement studies of the Web have focused on reference patterns established based on logs of proxies [9, 23], or servers [21] The authors in [4] captured client traces, but they concentrated on events at the user interface level in order to study browser and page design. In contrast, our goal in data collection was to acquire a complete picture of the reference behavior and timing of user accesses to ....

Jeff Sedayao. "Mosaic Will Kill My Network!" -- Studying Network Traffic Patterns of Mosaic Use. In Electronic Proceedings of the Second World Wide Web Conference '94: Mosaic and the Web, Chicago, Illinois, October 1994.

....experience, compound documents and the applications that operate on them need to adapt to the available bandwidth. To identify opportunities for adapting compound documents we need to understand their main characteristics. However, most studies of content types, especially those done on the Web [4, 21, 23, 24] have consistently ignored compound documents or treated them as opaque data streams, ignoring the rich internal structure that can be used to enhance bandwidth adaptation. In this paper we present an analysis of Oce compound documents downloaded from the Web. We focus on those characteristics of ....

....the ndings of section 4.2, where the size contribution of images to document size becomes the dominant factor as document size increases. The results for Word are similar, and are omitted for brevity. We compared the average size of images in Oce documents to the ndings of previous Web studies [2, 23]. In general, these studies report the average size of images between 5 KB and 22 KB. In comparison, Oce documents, especially PowerPoint documents, tend to have larger images. These results suggest that image distillation and other adaptation techniques are at least as important for compound ....

Sedayao, J. "Mosaic will kill my network!" - studying network trac patterns of Mosaic use. In Proc. of the 2nd International WWW Conference (Chicago, Illinois, 1994).

....to a proxy server) The characterization at this lower level is essential for estimating the traffic on the net and are thus the starting point for evaluations of net traffic. 1 Introduction The characterizaiton of internet WWW traffic at the system s level has been studied in several papers [2][3] 4] Our approach distinguishes from the others in that we try to map the actual user behavior (characterized in terms of the number of browser sessions a user would start and in terms of the number of conversations within a browser) to the traffic characteristics at the lower level. Thus, the ....

J. Sedayao, "Mosaic will kill my network!", in Proc. of Second International WWW Conference, 1994, http://www.ncsa.uiuc.edu/SDG/IT94/Proc."/DDay/sedayao/.

.... on Probabilistic Attributed Context Free Grammars (PACFG) Fu 74] which have proven to be a useful way to translate the workload from the application oriented user s point of view to the resource oriented system level in distributed environments ( Ragh 95] SV R 96] Previous work in this area [Seda 94] Arli 96] Brak 96] Cunh 95] Crov 96] has focused on the characterization of internet WWW traffic mainly at the system s level. Our approach distinguishes from the others in that we try to map the actual user behavior (characterized in terms of the number of browser sessions a user would start ....

J. Sedayao. "Mosaic Will Kill My Network!". In: Proceedings of Second International WWW Conference, Chicago, IL, USA, October 1994. http://www.ncsa.uiuc.edu/SDG/IT94/Proceedings/DDay/sedayao/.

....arrival rate is another parameter which allows for the dimension of cache size and number of servers. From the client side, request rate, preferences for subjects and files, and navigation strategies help in designing agents and browsers that reduce latency and are more friendly used. 38 Sedayao [111] was one of the first to start a study on Web traffic and on what a typical Mosaic user profile would be. His study was based on data collected at Intel using the logs of their proxy server and a network sniffer. He was motivated to find out how seriously Mosaic could hurt a company s network. His ....

Jeff Sedayao. "Mosaic Will Kill My Network!" -- Studying Network Traffic Patterns of Mosaic Use. In Electronic Proceedings of the Second World Wide Web Conference '94: Mosaic and the Web, Chicago, Illinois, October 1994.

....lack of response codes causes less than a 1 hit rate difference. 5 Related Work Since the Web became the primary consumer of Internet bandwidth, studies of Web traffic have become common. Some early studies include analysis of Web access traces from the perspective of browsers [6, 4] proxies [10, 20, 22], and servers [16] Arlitt et al. conducted a recent study of Web server workloads [2] Our work is unique in two ways. First, we examine many more requests, much larger caches, and much higher request rates; we also include data from many more sites. Second, unlike the earlier studies, we use a ....

Jeff Sedayao. "mosaic will kill my network!" - studying network traffic patterns of mosaic use. In Electronic Proceedings of the Second World Wide Web Conference '94: Mosaic and the Web, 1994.

....lack of response codes causes less than a 1 hit rate difference. 5 Related Work Since the Web became the primary consumer of Internet bandwidth, studies of Web traffic have become common. Some early studies include analysis of Web access traces from the perspective of browsers [6, 4] proxies [10, 20, 22], and servers [16] Arlitt et al. conducted a recent study of Web server workloads [2] Our work is unique in two ways. First, we examine many more requests, much larger caches, and much higher request rates; we also include data from many more sites. Second, unlike the earlier studies, we use a ....

Jeff Sedayao. "mosaic will kill my network!" - studying network traffic patterns of mosaic use. In Electronic Proceedings of the Second World Wide Web Conference '94: Mosaic and the Web, 1994.

....if the estimator settles to a consistent value, this value provides an estimate of ff. 3 Related Work The first step in understanding WWW traffic is the collection of trace data. Previous measurement studies of the Web have focused on reference patterns established based on logs of proxies [11, 25], or servers [21] The authors in [5] captured client traces, but they concentrated on events at the user interface level in order to study browser and page design. In contrast, our goal in data collection was to acquire a complete picture of the reference behavior and timing of user accesses to ....

Jeff Sedayao. "Mosaic will kill my network!" -- studying network traffic patterns of Mosaic use. In Electronic Proceedings of the Second World Wide Web Conference '94: Mosaic and the Web, Chicago, Illinois, October 1994.

....not the Other category. 28 Using Table 4.4, a second candidate invariant can be identified in Web server workloads. Across the six data sets, HTML and Image documents accounted for 90 100 of the total requests to the server. 3 This observation is consistent with results reported by Sedayao [60], by Cunha et al. 20] and by Kwan et al. 37] All of these papers reported that over 90 of client requests were for either HTML or Image documents. Table 4.4 also indicates that most transferred documents are quite small, which is a third invariant. This phenomenon was also observed by Braun ....

J. Sedayao, "Mosaic Will Kill My Network!", Electronic Proceedings of the Second World Wide Web Conference '94: Mosaic and the Web, Chicago, Illinois, October 1994. Available at URL: http://www.ncsa.uiuc.edu/SDG/IT94/Proceedings/DDay/ sedayao/mos traf paper.html

....lines, while the lognormal dataset is characterized by lines that seem roughly convergent. 3 Related Work The first step in understanding WWW traffic is the collection of trace data. Previous measurement studies of the Web have focused on reference patterns established based on logs of proxies [9, 23], or servers [21] The authors in [4] captured client traces, but they concentrated on events at the user interface level in order to study browser and page design. In contrast, our goal in data collection was to acquire a complete picture of the reference behavior and timing of user accesses to ....

Jeff Sedayao. "Mosaic Will Kill My Network!" -- Studying Network Traffic Patterns of Mosaic Use. In Electronic Proceedings of the Second World Wide Web Conference '94: Mosaic and the Web, Chicago, Illinois, October 1994.

....to the research community. The traces are available to the public via anonymous FTP. 1 2 Data Collection The first step in understanding patterns of WWW use is the collection of trace data. Previous file oriented studies have focused on reference patterns established based on logs of proxies [8, 14], or servers [13] The authors in [3] captured client traces, but they concentrated on events at the user interface level in order to study browser and page design. In contrast, our goal in data collection was to acquire a complete picture of the reference behavior and timing of user accesses to ....

Jeff Sedayao. "Mosaic Will Kill My Network!" -- Studying Network Traffic Patterns of Mosaic Use. In Electronic Proceedings of the Second World Wide Web Conference '94: Mosaic and the Web, Chicago, Illinois, October 1994.

....results for each log are given in Table 5. Using Table 5, we can identify a second invariant in Web server workloads. Across the six data sets, HTML and Image documents accounted for 90 100 of the total requests to the server. 4 This observation is consistent with results reported by Sedayao [19] and by Cunha, Bestavros and Crovella [7] Both of these papers reported that over 90 of client requests were for either HTML or image documents. Table 5 also indicates that most transferred documents are quite small, which is a third invariant. This phenomenon was also observed by Braun and ....

J. Sedayao, "Mosaic Will Kill My Network!", Electronic Proceedings of the Second World Wide Web Conference '94: Mosaic and the Web, Chicago, Illinois, October 1994.

....lines, while the lognormal dataset is characterized by lines that seem roughly convergent. 3 Related Work The first step in understanding WWW traffic is the collection of trace data. Previous measurement studies of the Web have focused on reference patterns established based on logs of proxies [10, 23], or servers [21] The authors in [5] captured client traces, but they concentrated on events at the user interface level in order to study browser and page design. In contrast, our goal in data collection was to acquire a complete picture of the reference behavior and timing of user accesses to ....

Jeff Sedayao. "Mosaic Will Kill My Network!" -- Studying Network Traffic Patterns of Mosaic Use. In Electronic Proceedings of the Second World Wide Web Conference '94: Mosaic and the Web, Chicago, Illinois, October 1994.

....process; next, the results of our simulations for various caching policies using that data; next, a comparison of our work with related research; and, finally, our conclusions. 2 Reference Patterns 2. 1 Data Collection Methods Prior studies of WWW traffic have been based on logs from proxies [7, 16], or logs from the HTTP server daemon [13] Our study required knowledge of individual user s access patterns, and we did not wish our data to be influenced by the caching behavior built in to the client application (Mosaic) For these reasons, we instrumented Mosaic directly and captured logs of ....

Jeff Sedayao. "Mosaic Will Kill My Network!" -- Studying Network Traffic Patterns of Mosaic Use. In Electronic Proc. of the 2nd WWW Conference, Chicago, Illinois, October 1994.

.... proposed mechanisms to reduce the latency associated with HTTP communications by removing the overhead of creating a new connection for every HTTP request [26] Sedayao studied the Mosaic related traffic on a large corporate network and presented several issues to consider when deploying Mosaic [30]. Pitkow and Recker have developed a dynamic caching algorithm for Web servers that is based on a model of human memory [27] In this paper we consider a technique that utilizes server to clients multicast communication to reduce the load placed on the server. At typical servers, requests for the ....

J. Sedayao. "Mosaic will kill my network!" - studying network traffic patterns of mosaic use. In Second World Wide Web Conference '94. http://- www.ncsa.uiuc.edu/SDG/IT94/Proceedings, October 1994.

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