×

Optimizing multi-hit caching for long tail content

  • US 8,639,780 B2
  • Filed: 12/19/2012
  • Issued: 01/28/2014
  • Est. Priority Date: 01/10/2012
  • Status: Active Grant
First Claim
Patent Images

1. A computer-implemented method performed by a caching server for N hit caching of content by that caching server, wherein N is an integer value greater than one, the computer-implemented method comprising:

  • configuring the caching server with a first request counter and a second request counter for tracking request counts for a plurality of content received during different intervals to different sets of indices of the first and second request counters, wherein each set of the different sets of indices uniquely identifies a request count for different content of the plurality of content;

    using the first request counter of the caching server to increment a request count for each content of a plurality of content that is requested during a first interval, wherein the first request counter tracks M requests received for a specific item of content of the plurality of content during the first interval to a particular set of indices, wherein M is an integer value that is less than N;

    resetting the first request counter at the end of the first interval, said resetting comprising (i) copying the request count tracked to the particular set of indices of the first request counter to the second request counter and (ii) clearing request counts of the first request counter;

    receiving at the caching server, a new request for the specific item of content during a second interval that immediately follows the first interval;

    incrementing the particular set of indices of the first request counter to track the new request for the specific item of content that is received during the second interval;

    determining a cumulative request count for the specific item of content based on the M requests from the first interval that were copied to the particular set of indices of the second request counter and at least the new request that is tracked to the particular set of indices of the first request counter during the second interval;

    passing the specific item of content from an origin and caching it at the caching server when the cumulative request count for the specific item of content identifies N requests; and

    passing the specific item of content from an origin without caching it at the caching server when the cumulative request count for the specific item of content identifies fewer than N requests.

View all claims
  • 6 Assignments
Timeline View
Assignment View
    ×
    ×