A. Iyengar, E. MacNair, and T. Nguyen. An analysis of web server performance. In Proceedings of GLOBECOM '97, 1997.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....on the foregoing non conventional performance metrics and workload characteristics. Some research studies have considered the issue of workloads with transient overload, but these studies have focused on single class workloads and specific scheduling strategies, such as admission control (e.g. [8]) and direct modifications to the Internet server scheduling mechanism (e.g. 2, 5] In contrast, our focus in this paper is on the optimal dynamic scheduling of a multiclass system with transient overload. Furthermore, very little research has even attempted to consider the issue of maximizing ....

A. Iyengar, E. MacNair and T. Nguyen. An analysis of web server performance. In Proceedings of GLOBECOM'97, 1997.

....or make extensions to the CPN model itself. In the literature, there are several papers on performance analysis of web servers. Many of these papers present measurement studies that focus on workload characterization [6, 14, 82] or measurement of, e.g. resource utilization and response time [2, 41, 68]. Analytic models have been used to analyze the performance of HTTP over several transport protocols [61] and for capacity planning of web servers [41] As one of the few simulation studies of web servers, 123] presents an end to end queueing model of a web server environment. Although these ....

A. Iyengar, E. MacNair, and T. Nguyen. An analysis of web server performance. In GLOBECOM 97, volume 3, pages 1943.

.... the server by a server farm (e.g. 8, 6, 16] Another solution is to enhance the OS to better support server software: 20, 9, 1, 15, 19] Finally, people try to limit the load at a server, either via installing a proxy cache (at the client or server end) e.g. 10, 4] or by admission control ([13]) Our approach to coping with transient overload is different from the above approaches. Our approach does not require buying more hardware or limiting the number of system users. We simply propose scheduling the jobs in a different order from that traditionally used. In computer systems today, ....

A. Iyengar, E. MacNair, and T. Nguyen. An analysis of web server performance. In Proceedings of GLOBECOM '97, 1997.

....Item Req. Trac ( Mean (KB) CoV HTML 19.2 15.0 5.76 1.90 Image 68.8 49.2 4.98 2.46 Audio 0.2 2.5 579.9 1.76 Video 0.1 6.7 2503.9 1.56 Dynamic 4.9 4.4 6.84 1.33 Other 6.8 20.2 19.0 7.90 Total 100 100 7.39 14.4 There are non negligible percentage of dynamic objects. Previous studies [18] have estimated that a dynamic object requires 10 to 100 times of service time than a typical static object. The processing time di erence was con rmed by recent benchmark tests of popular commercial web servers [8] and the following service time characteristic study. The throughput of serving ....

A. K. Iyengar, E. MacNair, and T. Nguyen. An analysis web server performance. In Proceedings of the IEEE 1997 Global Telecommunications Conference (GLOBECOM '97), Phoeniz, AZ, November 1997.

....Item Req. Trac ( Mean (KB) CoV HTML 19.2 15.0 5.76 1.90 Image 68.8 49.2 4.98 2.46 Audio 0.2 2.5 579.9 1.76 Video 0.1 6.7 2503.9 1.56 Dynamic 4.9 4.4 6.84 1.33 Other 6.8 20.2 19.0 7.90 Total 100 100 7.39 14.4 There are non negligible percentage of dynamic objects. Previous studies [18] have estimated that a dynamic object requires 10 to 100 times of service time than a typical static object. The processing time di erence was con rmed by recent benchmark tests of popular commercial web servers [8] and the following service time characteristic study. The throughput of serving ....

A. K. Iyengar, E. MacNair, and T. Nguyen. An analysis web server performance. In Proceedings of the IEEE 1997 Global Telecommunications Conference (GLOBECOM '97), Phoeniz, AZ, November 1997.

No context found.

A. Iyengar, E. MacNair, and T. Nguyen, "An analysis of Web server performance," in Proc. IEEE GLOBECOM, Nov. 1997.

....cached data. Consequently, we allow dynamic Web pages to be cached as well as static ones, since applications can explicitly invalidate any page whenever it becomes obsolete. Caching of dynamic Web pages is essential for improving the performance of Web sites containing significant dynamic content [2,3,14,15]. Multiprocessor accelerators can further increase the performance. Our multiprocessor system ar chitecture consists of a cluster of Web accelerator cache nodes and a front end load balancer. From a scalability standpoint, the objective is to combine the individual cache space of each member of ....

....data is typically serviced by returning a file. Web servers often consume several orders of magnitude more CPU time creating a dynamic page than a comparably sized static page. For Web sites containing significant dynamic content, it is essential to cache dynamic pages to improve performance [2,3,14,15]. We are not aware of any Web server accelerator besides our own which allows dynamic pages to be cached. All cached data must be stored in memory. Caching objects on disk would slow down the accelerator too much. Consequently, cache sizes are limited by memory sizes. Our accelerator uses the ....

A. Iyengar, E. MacNair, T. Nguyen, An analysis of web server performance, Proceedings of GLOBECOM'97, No- vember 1997.

....often orders of magnitude slower; it is not uncommon for a program to consume over a second of CPU time in order to generate a single dynamic page. For Web sites with a high proportion of dynamic pages, the performance bottleneck is often the CPU overhead associated with generating dynamic pages [6, 7]. Dynamic pages are essential at Web sites which provide data that change frequently. If pages are generated dynamically by a server program, the server program can return the most recent version of the data. If, on the other hand, files are created to serve the pages statically, it may not be ....

A. Iyengar, E. MacNair, and T. Nguyen. An Analysis of Web Server Performance. In Proceedings of GLOBECOM '97, November 1997.

....like the Olympic Games where Web pages are constantly changing. However, dynamic Web pages are often expensive to serve. A static page typically requires 2 to 10 milliseconds of CPU time to generate. By contrast, a dynamic page can consume several orders of magnitude more CPU time to generate [8]. In order to reduce the overhead for generating dynamic pages, our system attempts to cache dynamic pages when they are first created. Therefore, the overhead for creating a dynamic page is only incurred once. Subsequent requests for a cached dynamic page will obtain the page from the cache ....

A. Iyengar, E. MacNair, and T. Nguyen. An Analysis of Web Server Performance. In Proceedings of GLOBECOM '97, November 1997.

....at the time a request is made. The presence of dynamic data often slows down Web sites considerably. High performance Web servers can typically deliver several hundred static files per second. By contrast, the rate at which dynamic pages are delivered is often one or two order of magnitudes slower [10]. One technique for reducing the overhead of dynamic page creation is to cache dynamic pages at the server the first time they are created. That way, subsequent requests for the same dynamic page can access the page from the cache instead of repeatedly invoking a program to generate the same page. ....

A. Iyengar, E. MacNair, and T. Nguyen. An Analysis of Web Server Performance. In Proceedings of GLOBECOM '97, November 1997.

....like the Olympic Games where Web pages are constantly changing. However, dynamic Web pages are often expensive to serve. A static page typically requires 2 to 10 milliseconds of CPU time to generate. By contrast, a dynamic page can consume several orders of magnitude more CPU time to generate [8]. In order to reduce the overhead for generating dynamic pages, our system attempts to cache dynamic pages when they are first created. Therefore, the overhead for creating a dynamic page is only incurred once. Subsequent requests for a cached dynamic page will obtain the page from the cache ....

A. Iyengar, E. MacNair, and T. Nguyen. An Analysis of Web Server Performance. In Proceedings of GLOBECOM '97, November 1997.

....cached data. Consequently, we allow dynamic Web pages to be cached as well as static ones, since applications can explicitly invalidate any page whenever it becomes obsolete. Caching of dynamic Web pages is essential for improving the performance of Web sites containing significant dynamic content [10, 9, 4, 3]. Httpd accelerators are contained in both the Harvest and Squid caches [5, 14] Our httpd accelerator results in considerably better performance than the Harvest and Squid accelerators partly because our accelerator runs on an embedded operating system. Novell sells an httpd accelerator as part ....

....often makes it feasible to cache dynamic Web pages. Web servers often consume several orders of magnitude more CPU time creating a dynamic page than a comparably sized static page. For Web sites containing significant dynamic content, it is essential to cache dynamic pages to improve performance [10, 9, 4, 3]. We are not aware of any httpd accelerator besides our own which allows dynamic pages to be cached. All cached data must be stored in memory. Caching objects on disk would slow down the accelerator too much. Consequently, cache sizes are limited by memory sizes. Our accelerator uses the least ....

A. Iyengar, E. MacNair, and T. Nguyen. An Analysis of Web Server Performance. In Proceedings of GLOBECOM '97, November 1997.

....often orders of magnitude slower; it is not uncommon for a program to consume over a second of CPU time in order to generate a single dynamic page. For Web sites with a high proportion of dynamic pages, the performance bottleneck is often the CPU overhead associated with generating dynamic pages [6, 7]. Dynamic pages are essential at Web sites which provide data that change frequently. If pages are generated dynamically by a server program, the server program can return the most recent version of the data. If, on the other hand, files are created to serve the pages statically, it may not be ....

A. Iyengar, E. MacNair, and T. Nguyen. An Analysis of Web Server Performance. In Proceedings of GLOBECOM '97, November 1997.

....A critical issue for continued and successful growth concerns the performance of Web servers, which must provide reliable, scalable and efficient access to Internet services. A significant body of work in the research literature has focused on various aspects of Web server performance; e.g. see [7, 8, 10] and the references therein. Crovella and Bestavros have found evidence that Web traffic is selfsimilar [6] Arlitt and Williamson identified ten workload invariants by analyzing logs from six different Web servers [2] Barford and Crovella have developed a tool that generates workloads based on ....

....of 9.52 requests per second. 0 5 10 15 20 25 30 0 500 1000 1500 2000 2500 3000 3500 Number of Requests Per Second Time (seconds) ARMA(1,2) Process, Mean=9.52 Number of Requests Figure 2. Scaled ARMA(1,2) process. An alternative method, which we call M2M, has been used in other studies (e.g. [10]) to analyze and scale Web server access distributions. Specifically, the Web server request distributions were characterized by the interrequest time mean and cv A as calculated from the Web access logs. In order to scale an interrequest time distribution to an arbitrary request rate, there is ....

[Article contains additional citation context not shown here]

A. Iyengar, E. MacNair, and T. Nguyen. An Analysis of Web Server Performance. In Proceedings of GLOBECOM '97, 1997.

No context found.

A. Iyengar, E. MacNair, and T. Nguyen. An analysis of web server performance. In Proceedings of GLOBECOM '97, 1997.

No context found.

A. Iyengar, E. MacNair and T. Nguyen, "An analysis of web server performance", Proc. of Globecom'97,

No context found.

A. Iyengar, E. MacNair, and T. Nguyen, "An Analysis of Web Server Performance, " in Proceedings of GLOBECOM '97, 1997.

No context found.

A. Iyengar, E. MacNair, and T. Nguyen. An Analysis of Web Server Performance. In Proceedings of GLOBECOM '97,November 1997.

No context found.

A. Iyengar, E. MacNair and T. Nguyen (1997). "An Analysis of Web Server Performance." Proceedings of GLOBECOM '97. http://citeseer.nj.nec.com/83344.html

No context found.

A. Iyengar, E. MacNair, and T. Nguyen. An analysis of web server performance. In Proceedings of GLOBECOM '97, 1997.

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