E. Nahum, M. Rosu, S. Seshan, and J. Almeida. The effects of wide-area conditions on WWW server performance. In Proc of ACM SIGMETRICS, pages 257--267, 2001.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....These studies typically concentrate on the effect that network protocols conditions have on web server performance. We list just a few below: In [4] and [14] the authors find that the TCP RTO value has a large impact on server performance under FreeBSD. This agrees with our study. In [35] the authors study the effect of WAN conditions, and find that losses and delays can affect response times. They use a different workload from ours (Surge workload) but have similar findings. The benefits of persistent connections are evaluated by [32] and [10] in a LAN environment. There are ....

E. Nahum, M. Rosu, S. Seshan, and J. Almeida. The effects of widearea conditions on www server performance. In Proceedings of ACM SIGMETRICS '01, pages 257--267, 2001.

.... addressed viable CDN architectures [48, 8, 4, 33, 12, 43, 37, 16] the role of a proxy in a CDN [32] cooperative caching to improve hit ratio and response times [44, 5, 11, 43, 12, 17, 28, 34, 38, 49] load balancing amongst proxies [23, 24, 33] redirection schemes and other performance issues [21, 20, 31], object replication strategies in such networks [22] and how prefetching affects performance in CDNs [42] very few have addressed consistency issues in CDNs. 47, 46, 48, 45, 33] are few of the existing studies that have looked into the area of managing consistency in large scale systems. While ....

E. Nahum, M. Rosu, S. Seshan, and J. Almeida. The effects of wide-area conditions on www server performance. In ACM SIGMETRICS 2001.

....Bernoulli case, each packet traversing a link is dropped with a fixed probability determined by the loss rate of the link. In the Gilbert case, the link fluctuates between a good state and a bad state. In the good state, no packets are dropped while in the bad state all packets are dropped. As in [13], we chose the probability of remaining in the bad state to be 35 based on Paxson s observed measurements of the Internet. The other state transition probabilities are picked so that the average loss rate matches the loss rate assigned to the link. Thus, the Gilbert loss process is likely to ....

E. M. Nahum, C.-C. Rosu, S. Seshan, and J. Almeida. The Effects of Wide-Area Conditions on WWW Server Performance. In Proceedings of ACM SIGMETRICS, June 2001.

....Bernoulli case, each packet traversing a link is dropped with a fixed probability determined by the loss rate of the link. In the Gilbert case, the link fluctuates between a good state and a bad state. At the good state, no packets are dropped while at the bad state all packets are dropped. As in [16], we chose the probability of remaining in the bad state to be 35 based on Paxson s observed measurements of the Internet. The other state transition probability is picked so that the average loss rate matches the loss rate assigned to the link. Thus the Gilbert loss process is likely to generate ....

E. M. Nahum, C.-C. Rosu, S. Seshan, and J. Almeida. The effects of wide-area conditions on www server performance. In Proceedings of ACM SIGMETRICS, June 2001.

....to resume an earlier TLS session and thus reuse the result of an earlier RSA computation. Research has suggested that, indeed, session caching helps web server performance [11] Likewise, there has been considerable prior work in performance analysis and benchmarking of conventional web servers [15, 12, 17, 5, 18], performance optimizations of web servers, performance oriented web server design, and operating system support for web servers [13, 22, 6, 7, 21] Apostolopuolos et al. 3] studied the cost of TLS connection setup, RC4 and MD5, and proposed TLS connection setup protocol changes. Our methodology ....

E. M. Nahum, M. Rosu, S. Seshan, and J. Almeida. The effects of wide-area conditions on WWW server performance. In Proceedings of the ACM SIGMETRICS Conference on Measurement and Modeling of Computer Systems, Cambridge, Massachusetts, June 2001.

....limits to packet flows between hosts in the local area. Two tools which enable WAN emulation are available at present: dummynet [33] and NISTnet [128] While these tools have been available for some time, only recently have studies been made about their use in Web workload characterization. In [89], Nahum et al. use dummynet to investigate the effects of WAN emulation in local area web server performance tests. That work focuses on comparing the performance of the Reno and SACK versions of TCP. We utilize the NISTnet WAN emulator in our work and present the results of server performance ....

E. Nahum, J. Almeida, M. Rosu, and S. Seshan. The effects of wide-area conditions on WWW server performance. Currently under review for publication, November 2000.

No context found.

NAHUM, E. M., ROSU, M., SESHAN, S., AND ALMEIDA, J. The Effects of Wide-Area Conditions on WWW Server Performance. In Proc. ACM SIGMETRICS (Cambridge, MA, June 2001).

No context found.

E. Nahum, M. Rosu, S. Seshan, J. Almeida, The Effects of Wide Area Conditions on WWW Server Performance, SIGMETRICS, 2001.

....in user space. 2.5. Server Performance Comparison Since we are concerned with performance, it is thus interesting to see how well the different server architectures perform. To evaluate them, we took a experimental testbed setup and evaluate the performance using a synthetic workload generator [51] to saturate the servers with requests for a range of web documents. The clients were eight 500 MHz PC s running FreeBSD, and the server was a 400 MHz PC running Linux 2.4.16. Each client had a 100 mbps Ethernet connected to a gigabit switch, and the server was connected to the switch using ....

E. M. Nahum, M. Rosu, S. Seshan, and J. Almeida. The effects of wide-area conditions on WWW server performance. In Proceedings of the ACM Sigmetrics Conference on Measurement and Modeling of Computer Systems, Cambridge, MA, June 2001.

....pieces of hardware, or to compare two different operating systems running on the same hardware. Many Web server benchmarks have appeared over the years, including WebStone [34] SPECweb96 [15] S Client [7] SURGE [8] httperf [29] SPECweb99 [15] TPC W [33] WAGON [26] and WaspClient [30]. As people s understanding of Web server behavior has expanded over time, so have the server benchmarks evolved and been improved. In addition to providing broad performance statistics, such as throughput in HTTP operations second, many benchmarks attempt to ad Responses Requests Server ....

....[9] httperf [29] is a workload generator which has the main goal of extensibility and broad correctness testing. httperf allows a wide variety of options and configurations to probe a range of scenarios on the server. httperf also includes the open loop model used by S Client. WaspClient [30] is a hybrid workload generator built by using the workload model from SURGE with the request generation engine from S Client. WaspClient was used to evaluate server performance in wide area conditions, rather than over a LAN. TPC W [33] is a specification for generating a workload for a full Web ....

Erich M. Nahum, Marcel Rosu, Srinivasan Seshan, and Jussara Almeida. The effects of widearea conditions on WWW server performance. In Proceedings of the ACM Sigmetrics Conference on Cambridge, MA, June 2001.

No context found.

E. Nahum, M. Rosu, S. Seshan, and J. Almeida. The effects of wide-area conditions on WWW server performance. In Proc of ACM SIGMETRICS, pages 257--267, 2001.

No context found.

E. M. Nahum, M.-C. Rosu, S. Seshan, and J. Almeida. The effects of wide-area conditions on www server performance. In Proceedings of ACM SIGMETRICS, June 2001.

No context found.

E. Nahum, M. Rosu, S. Seshan, and J. Almeida. The effects of widearea conditions on www server performance. In Proceedings of ACM SIGMETRICS '01, pages 257--267, 2001.

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