System and method for populating cache servers with popular media contents
AssigneeIntel Corporation
InventorFarid Adrangi
Life of this patent
Estimating the term…
Abstract
A system and method for caching files is disclosed. Popularity values are calculated for a plurality of files over a period of time. The popularity values are then used to determine which files should be cached at various remote sites. Once the caches are filled, the popularity values associated with the cached files are periodically compared with the popularity values of uncached content. Generally, if the popularity of an uncached file is greater than the popularity of a cached file, then the cached file will be replaced. However, numerous different variables may be factored into the caching determination including, for example, the size of the file, the required bitrate of the file, the identity of the owner of the file, and/or the type of file.
Claims29
A method for network caching comprising:
- calculating popularity values for a plurality of network files, said popularity values based on a number of requests for each of said files over a period of time;
- and
- transferring a subset of said plurality of network files to a cache based on said calculated popularity values.
A system for caching files comprising:
- one or more popularity calculation modules to calculate popularity values for a plurality of files based at least in part on a number of user requests for each of said plurality of files and to generate a prioritized list of said plurality of files based on said popularity values;
- and
- one or more caching analysis modules to identify one or more of said plurality of files to be cached based on said prioritized list;
- and
- one or more file transfer modules to transfer said one or more of said plurality of files to a cache.
An article of manufacture including a sequence of instructions which, when executed by a processor, causes said processor to:
- calculate popularity values for a plurality of network files, said popularity values based on a number of requests for each of said files over a period of time;
- and
- cache a subset of said plurality of network files based on said calculated popularity values.
Dependencies are inferred from the claim text.