Electrical computers and digital processing systems: multicomput – Computer network managing – Computer network monitoring
Reexamination Certificate
1998-02-25
2002-08-20
Meky, Moustafa M. (Department: 2757)
Electrical computers and digital processing systems: multicomput
Computer network managing
Computer network monitoring
C709S203000, C709S217000
Reexamination Certificate
active
06438592
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to a system for monitoring and improving performance of client-server hypermedia, such as, for example, the downloading of pages and page components on the World Wide Web.
BACKGROUND OF THE INVENTION
As the computer revolution advances, computer networking has become increasingly important. In recent years the number of computers which are connected to computer networks has increased rapidly. Not only are computers being connected with local networks, which might exist in a given building or group of buildings, but also with wide area networks, which commonly connect local area networks in widely separated locations, such as the different facilities of a large corporation. In fact, within the last several years it has become increasingly common for computers to be hooked up to a global network formed of a large number of sub-networks called the Internet.
An increasingly important use of networking is for distributing client-server networked hypermedia. In such a use, a server computer responds to a request from a client computer over an internetwork for a specific media objects, such as a World Wide Web page. It responds by sending the requested media object to the requesting client computer. The request message identifies the network address of the client computer which has sent it, the address of the server computer to which it is addressed and the address of the desired media object. The address of the desired media object is identified by a location specification which identifies both the address of the server computer to which the request is addressed and the address within the storage system of the server of the desired object. In the World Wide Web such an identifier is called Uniform Resource Locator, or URL.
Many requested media objects are composite objects. These contain within their data location specifications, such as URLs, which link to other, component, media objects, such as image, audio, video, or programming objects. They are automatically accessed and used in the display of their associated composite objects. For example, most World Wide Web pages include URL links to image files which are automatically requested and displayed as part of the such pages. In addition to such automatically called links, a media object, such as a World Wide Web page, can include links to media objects which are only requested if a user selects to have them called, such as by clicking a mouse on a portion of a computer screen associated with such a link.
The present invention relates to such client-server hypermedia. The World Wide Web is currently by far the most important example of such hypermedia, but in this specification and the claims when I use the word “Web” I mean not only to refer to the World Wide Web itself, as it currently exists, but also future forms of such client-server hypermedia, such as that which might exist on a second generation Internet that is currently being planned. Similarly, when I refer to URLs in the claims that follow I intend to include other location specifications which might be used in client-server hypertext media of the future.
It is common for those who distribute information and services over the Web through a server system to want to monitor the level of performance they are providing to users and, if possible, to modify the operation of the server so as to improve that performance.
In the prior art there have been multiple systems which enable operators of Web servers to chart information contained in URL requests which the server has received from client computers. This information includes such statistics as distribution of source addresses of client computers from which such requests have been received and the distribution of files which have been requested.
The prior art has also included server systems which monitor the performance of the server computer itself. For example, the prior art has includes system which can provide an indication of the degree of busyness of a server. Such systems can provide those who operate Web servers with valuable information about the number of URL requests the server is receiving and the number of messages or bytes a server is sending out at any point in time. They can also indicate the time delay between the receipt of a request for a given media object and the time that the object is sent out.
Such measurements have been used in so-called IP spreading servers. Such servers are comprised of at least one spreader computer and two or more backend servers connected by the Internet. The spreader computer distributes the work of responding to URL requests between its associated backend servers as a function of how busy each individual backend server is. In such IP spreading servers, URL links to the server are all addressed to the spreader computer. When the spreader receives such requests from a client, it selects which of its backend servers should handle the request. It then sends the URL request to the selected backend server by placing it on the Internet with the selected backend server's Internet Protocol (“IP”) address as the request's destination address. The spreader varies from normal Internet Protocol by using the source IP address of client which sent the URL request to the spreader rather than the spreader's own IP address as the relayed request's source address. When the backend server receives the relayed request, it places one or more messages containing the contents of the file identified in the request's URL on the Internet. Each of these response messages is addressed to the client IP address identified by the source address in the relayed URL request received by the backend server. Like the spreader, the backend server varies from normal Internet Protocol by sending such response messages out with other than the backend server's own address as the source address. Instead, it uses the spreader computer's address as the source address for such response messages, since the client is expecting a response from the same address to which it sent the original URL request.
Unfortunately, monitoring performance on server systems does not give complete information about how long it actually takes a customer to display a given composite Web page. The display of the Web page normally only comes together at the client. The time to download a composite page includes the time required to download multiple different objects, and to my best knowledge no current servers determine which requests for Web page component are associated with which requests for Web pages. Monitoring performance on a server also fails to provide an accurate picture of the performance witnessed by clients because network delays external to the server can affect the performance observed by a client. Furthermore, when a server gets severely backed up, there often becomes a significant delay before a server can even accept receipt of a request, and there is no method for a server to measure delays which occur before it accepts receipt of a message.
To address the issue of monitoring how long it takes for Web objects to download onto client computers, the prior art has developed the technique of “mimicing”. This involves using one or more client computers located at various points on the Internet which have been programmed to measure the amount of time required to download a Web page and to provide a report on such measurements. Unfortunately this technique is spotty at best, because usually the number of mimicing computers, and the number of downloads to such computers, is extremely small compared to the number of clients and downloads a large server might have. Furthermore, mimicing is only a mechanical attempt to imitate the use of the server by its human users. As a result, it often fails to accurately monitor the performance actual human users of the server see.
SUMMARY OF THE INVENTION
It is an object of the present invention to provide apparatuses and methods for improving the ability of those who operate Web servers to monitor the ser
Meky Moustafa M.
Porter Edward W.
LandOfFree
Systems for monitoring and improving performance on the... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Systems for monitoring and improving performance on the..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Systems for monitoring and improving performance on the... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2920148