Method and system for caching streaming multimedia on the internet
View Patent ↗An apparatus and method to enhance existing caches in a network to better support streaming media storage and distribution. Helper machines are used inside the network to implement several methods which support streaming media including segmentation of streaming media objects into smaller units, cooperation of Helper machines, and novel placement and replacement policies for segments of media objects.
1. A method for caching streaming multimedia (SM), comprising:
calculating a helper hotness rating at a helper server (HS) for each of a plurality of SM objects that are hosted at a content server connected to the HS in a network, each helper hotness rating being a total number of client requests for a particular SM object divided, by a time period during which the client requests are received;
categorizing each SM object into one of a plurality of helper hotness categories based on the calculated helper hotness rating; and
pulling, for each SM object, a fraction of the SM object by the HS from the content server, said fraction being determined according to the helper hotness category, the HS caching the fractions of each SM object for distribution to a plurality of clients, the HS being interposed between the content server and the clients.
2. The method of claim 1 , further comprising:
calculating a helper hotness rating at each of a plurality of helper servers (HSs) for each of a plurality of SM objects that are hosted at a content server connected to the HSs in a network, each helper hotness rating being a total number of client requests for a particular SM object divided by a time period during which the client requests are received;
calculating a server hotness rating at the content server for each SM object, each server hotness rating for the particular SM object being a sum of the calculated helper hotness ratings received from all of the HSs;
categorizing each SM object into one of a plurality of server hotness categories based on the calculated server hotness rating; and
pushing each SM object from the content server to a fraction of HSs, each fraction being determined according to the server hotness category, the HSs caching the fractions of each SM object for distribution to a plurality of clients, the HSs being interposed between the content server and the clients.
3. The method as recited in claim 1 , wherein pulling by said HS is in response to client requests.
4. The method as recited in claim 1 , wherein a deterministic cache placement and replacement policy is implemented at the HSs.
5. The method of claim 1 , wherein a random cache placement and replacement policy is implemented at the HSs.
6. The method as recited in claim 2 , wherein pushing by said content server is in response to client requests.
7. The method as recited in claim 2 , wherein a deterministic cache placement and replacement policy is implemented at the HSs.
8. The method of claim 2 , wherein a random cache placement and replacement policy is implemented at the HSs.