Why does it take so long to download a Web page from a Web server? We analyze the download latency for pages for a variety of situations, in which Web browser and server are both within the same country as well as in different countries. Our study examines several sources of latency in accessing Web pages: DNS, TCP, the Web server itself, and the network links and routers. We divide the total download time into four parts: DNS query, connection setup time, time to get the first byte of a Web page, and downloading time. In most cases, roughly half of the time is spent from the moment the browser sends the acknowledgement completing the TCP connection establishment until the first packet containing page content arrives. The bulk of this time is the round trip delay, and only a tiny portion is delay at the server. This implies that the bottleneck in accessing pages over the Internet is due to the Internet itself, and not the server speed (as suggested by another study). The second bottleneck is the 3-way TCP connection establishment (consuming 1/5 to 1/4 of the delay). Conclusions are drawn on how to decrease latency.

Abstract. Sampling has become an integral part of passive network measurement. This role is driven by the need to control the consumption of resources in the measurement infrastructure under increasing traffic rates and the demand for detailed measurements from applications and service providers. Classical sampling methods play an important role in the current practice of Internet measurement. The aims of this review are (i) to explain the classical sampling methodology in the context of the Internet to readers who are not necessarily acquainted with either, (ii) to give an account of newer applications and sampling methods for passive measurement and (iii) to identify emerging areas that are ripe for the application of statistical expertise. Key words and phrases: Traffic measurement, network management, sampling methods, estimation, packets, flows.

A major source of performance degradation in highperformance distributed applications has been attributed to poor end-to-end TCP performance. The root causes of poor TCP performance are difficult to isolate and diagnose, and the efficacy of tuning efforts are often difficult to gauge. This paper describes some sources of poor TCP performance, and describes a method to diagnose some of these problems based on a combination of existing performance tools and the Web100 tuning package. Using this methodology, the TCP performance of an application developed for the Visible Human project is shown to significantly improve. 1

A major source of performance degradation in highperformance distributed applications has been attributed to poor end-to-end Transport Control Protocol (TCP) performance. The root causes of poor TCP performance are difficult to isolate and diagnose, and the efficacy of tuning efforts are often difficult to gauge. This paper describes sources of poor TCP performance, and suggests a method to diagnose some of these problems based on a combination of existing performance tools and the Web100 tuning package. Using this methodology, the TCP performance of an application developed for the Visible Human Project is shown to significantly improve, making it practical for full resolution Visible Human data to be used

Monitoring in a grid environment involves the analysis of all resource metrics and network metrics. Monitoring the resource metrics helps the grid middleware to decide which job to be submitted to which resource. Decision for submitting a job will be better along with the consideration of the network metrics. Tuning of the network metrics is also made if the performance degrades.