Electrical computers and digital processing systems: multicomput – Computer-to-computer data routing – Least weight routing
Reexamination Certificate
2000-07-03
2004-08-31
Harvey, Jack B. (Department: 2142)
Electrical computers and digital processing systems: multicomput
Computer-to-computer data routing
Least weight routing
C709S201000, C709S203000, C709S225000, C709S238000, C709S239000, C709S245000
Reexamination Certificate
active
06785704
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention relates to efficient transmission of data in an internetwork, such as the global internetwork known as the “Internet”. More specifically, the present invention relates to moving live or stored “broadcast” data streams from content producers to large numbers of recipients of those data streams.
“Broadcasting” refers to the transmission of a data stream from a content producer to a large number of recipients. The data stream can be text, graphics, video, audio, or any other digital data stream. Data is often provided as a stream or as a file, with the distinction being that the end of the stream is open-ended, while the file has a defined end. For example, real-time stock quotes might be thought of as a stream of data, while a 30-minute audiovisual presentation might be thought of as a file of data. As referenced herein, a sharp distinction is not needed between what is a stream and what is a file, since the typical broadcast operation is very similar whether a stream is being transmitted or a file is being transmitted. Therefore, it should be understood that where a stream is described, a file could be substituted therefor unless otherwise indicated.
Broadcasting need not be done in real-time relative to the content creation. Real-time broadcasting refers to the transmission of the data as it is created in a digital form. For example, a football game might be recorded by a camera, digitized and broadcast to many individuals wanting to receive the transmission over the Internet. The football game might also be stored after digitization and broadcast at a later time. Furthermore, the football game could be both transmitted live and transmitted at a later time (“delayed broadcast”). Generally speaking, whether the broadcast is live or delayed, some of the components of a broadcast network might operate exactly the same, as is the case with current television broadcasting. For example, the antennas broadcasting the signal and the receivers receiving the signal operate identically to receive live broadcasts or delayed broadcasts.
One technical difference between live broadcasting and delayed broadcasting is that live broadcasting is likely to have a larger audience at the time of the broadcast, since there is only one time to tune into a live broadcast but many times might be available for a delayed broadcast. Some content is more likely to be desired by recipients as a live broadcast rather than a delayed broadcast. Examples include sporting events, time-sensitive business information such as stock quotes, analyst interviews and breaking news, and the like.
The line between live and delayed broadcasting is not a fixed line. One of the challenges of live broadcasting is to process the data stream in real-time to make it suitable for transmission (e.g., compression, formatting), whereas more time is available for those processing steps if the data stream is a delayed broadcast. While that challenge highlights a distinction between live and delayed broadcasting, if the delayed broadcasts are only available at set times, as is the case with television reruns, live and delayed broadcasting do not differ greatly. Because the line is not always clear, it should be understood that “broadcasting” refers to live and/or delayed broadcasting unless otherwise indicated.
Relative to the current demands of Internet users, current television broadcasting is simple: the content creators provide their content to the broadcasters, who send out the data stream into a channel that is exclusively reserved for their content and has the bandwidth to carry that content in the time allotted and the medium of transfer (being wired or wireless) and the receivers are all connected to the medium with a bandwidth sufficient to receive all the data stream with minimal processing from a channel dedicated to that content. On the other hand, broadcasting content over the Internet (or any other internetwork or network being used) cannot be done easily, as the Internet or network is essentially a point to point transmission medium, with some provision for point-to-multipoint or multipoint-to-multipoint transmission.
For example, broadcast television deals with a breaking news event by gathering information, writing a script and putting a reporter on the air. The recipients of the breaking news (the television watchers) must wait for the broadcast television station to broadcast the information and they get only the data stream that that content provider chooses to present. When news breaks on the Internet, a large number of users will try to retrieve the news information (essentially as a large number of point-to-point transmissions of the same data stream), often swamping the servers and computing infrastructure of the content provider. This “flash effect” is not limited to breaking news, but is often encountered when live events occur, when new releases of popular software are issued or when a popular Web site is encountered. Herein, a “Web site” generally refers to a collection of pages presented as a unit, usually served from one or more coordinated servers having a particular network address and may also refer to the computers and infrastructure that serve the pages of the collection.
The problems of current broadcasting approaches are described below, but first some background of client-server architecture is in order. Many networking and other computing systems have the processing and functionality of the overall system separated into “clients” and “servers” where the clients are computers, programs or hardware that initiate requests and servers are computers, programs or hardware that respond to requests from clients. There are exceptions, where devices or programs generally thought of as servers will make requests of devices or programs generally thought of as clients, but for the most part, in the client-server model, the servers wait around for requests, service the requests and then wait around for further requests. Clients are usually considered more independent actors, in that they initiate requests. It should be understood, however, that some devices or hardware could be clients at some times or for some purposes and servers at other times or for other purposes.
In the context of an extremely basic broadcast infrastructure, a content server waits for a request from a content client and upon receipt of a request sends the requested content to the content client. This basic infrastructure is fine when one client makes a request of one server and the content fits in unused bandwidth of a channel connecting the client and server, but since most networks have more than one client or more that one server and share a limited bandwidth, the bandwidth needs to be intelligently allocated.
From an infrastructure perspective, the flash effect is not very bandwidth-efficient, as many, many identical copies of the data stream are transported over the network to the many recipients requesting the data stream. This effect might not be a problem if the data stream is a few bits of data, but data streams of full-motion video and CD-quality audio are becoming more and more common.
Several different approaches have been made in the past to provide for broadcasting over the Internet, but most have drawbacks that prevent their widespread adoption. Two key mechanisms for the Internet have been proposed and are in limited used to overcome the problems induced by the flash effect, namely, 1) caching and 2) server replication. Caching refers to a process of using a cache situated at strategic locations within the network infrastructure to intercept content requests from clients so that the content source does not need to provide every copy of the content. When a client requests content from a content server and the client receives the content from the content server, a cache in the network through which the content passes stores a copy of the content. When other client (or the same client) makes a request for that same content, the network infrastructure consults the cache to de
Albert Philip H.
FastForward Networks
Harvey Jack B.
Le Hieu C.
Townsend and Townsend / and Crew LLP
LandOfFree
Content distribution system for operation over an... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Content distribution system for operation over an..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Content distribution system for operation over an... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3283254