Browse Prior Art Database

Real-time measurement of Web Content Access performance Disclosure Number: IPCOM000004713D
Original Publication Date: 2001-Apr-17
Included in the Prior Art Database: 2001-Apr-17
Document File: 4 page(s) / 82K

Publishing Venue


Related People

Janusz Hyziak: AUTHOR [+2]


Real-time measurement of Web Content Access performance

This text was extracted from a WORD97 document.
This is the abbreviated version, containing approximately 36% of the total text.

Real-time measurement of Web Content Access performance

By: Janusz Hyziak, Shreesha Ramanna, and Kurian Jacob


For Web content searches, the user selects from mirror content servers, which are typically presented for a user to select, at the user's discretion. Many technology-enabling companies prosper in the Web-centric marketplace by providing solutions that optimize Web site performance, and provide interactive application services to determine the best physical architecture and optimal location of content servers. The problem is to determine which site is the best for a given user based on their access point to the network, and conditions in the network. Existing Web searches provide a Web server with a database of links arranged with respect to some key element/criteria. However these searches do not address any optimized path/ weights associated with each link. The approach provided in this paper offers a unique opportunity for the requestor to get up-to-date views of the content server loading and route conditions, to choose the optimal source of the content, including the delivery path for the content. Optionally, the search engine itself could provide the best content server to be selected based on the users profile, preferences, time-of-day and other parameters.


Most commonly, finding information on the Web is typically done through search engines. When the requestor specifies a search criterion, the search engines will retrieve the link to the information related to the search. The information returned to the requestor is a list of URLs and does not include any information about the accessibility or retrieval speed of that information. When the user requests retrieval from an address, the user has no idea how efficient the transport is or how long it will take to retrieve the information. The retrieval may not be the best, in terms of optimum path or least expensive route, to reach a destination.

This approach makes information available to the requestor such that, it provides a unique opportunity for the requestor to fetch the information from the nearest point, the least expensive route, or the fastest route. Most importantly, the traffic patterns along any given path can change with the time of day and this drastically affects the transmission times.

Having a method to measure the performance of the paths from the content server to the requesting user and to use the information for retrieval will make the best use of the network resources and improve satisfaction.


Figure 1 illustrates the current scheme. When the requestor issues a request for

information retrieval from the search engine the search engine returns both the destination addresses (in our case Server A and Server B).

The proposal is to provide qualifiers like shortest path, least expensive, least traffic, shortest delay, fastest transm...