Granted Patent B2

US 9,712,854 · App. 13/604,756 · Granted Jul 18, 2017

Abstract

Claims (34)

1. An apparatus, comprising:

a processor and a memory communicatively connected to the processor, the processor configured to:

determine a fraction of a plurality of content item versions to cache at a first network data storage location based on a popularity distribution of the content item versions and based on cost model information associated with the first network data storage location and a second network data storage location, wherein the cost model information comprises at least one of:

a first set of cost model information comprising a cost of storing one of the content item versions at the first network data storage location and a cost of transcoding one of the content item versions at the second network data storage location, wherein the first network data storage location is located at a first hierarchical level and the second network data storage location is located at a second hierarchical level, wherein the first hierarchical level is located farther from end users than the second hierarchical level; or

a second set of cost model information comprising a cost of storing one of the content item versions at the first network data storage location and a cost of transferring one of the content item versions from the second network data storage location to the first network data storage location, wherein the first network data storage location is located at a first hierarchical level and the second network data storage location is located at a second hierarchical level, wherein the first hierarchical level is located closer to end users than the second hierarchical level.

2. The apparatus of claim 1 , wherein the popularity distribution of the content item versions is based on popularity of the content item versions at the first network data storage location.

3. The apparatus of claim 1 , wherein, for the first set of cost model information, the first network data storage location comprises a centralized data center and the second network data storage location comprises an edge data center.

4. The apparatus of claim 1 , wherein, for the first set of cost model information, the processor is configured to determine the fraction of content item versions to cache at the first network data storage location based on eta=10^((log 10 (Etranscode/(Etranscode+Cstorage)))/α), wherein:

eta is the fraction of content item versions to cache at the first network data storage location;

Cstorage is the cost of storing one of the content item versions at the first network data storage location;

Etranscode is the cost of transcoding one of the content item versions at the second network data storage location; and

α is a skewness factor associated with the popularity distribution of the content item versions accessed at the first network data storage location.

5. The apparatus of claim 1 , wherein, for the second set of cost model information, the first network data storage location comprises an edge data center and the second network data storage location comprises a centralized data center.

6. The apparatus of claim 1 , wherein, for the second set of cost model information, the processor is configured to determine the fraction of content item versions to cache at the first network data storage location based on eta=10^((log 10 (Cbandwidth/(Cbandwidth+Estorage)))/α), wherein:

eta is the fraction of content item versions to cache at the first network data storage location;

Estorage is the cost of storing one of the content item versions at the first network data storage location;

Cbandwidth is the cost of transferring one of the content item versions from the second network data storage location to the first network data storage location; and

α is a skewness factor associated with the popularity distribution of the content item versions accessed at the first network data storage location.

7. The apparatus of claim 1 , wherein the popularity distribution of the content item versions has a skewness factor associated therewith, wherein the fraction of the content items to cache at the first network data storage location is determined based on the skewness factor.

8. The apparatus of claim 1 , wherein, to determine the fraction of content item versions to cache at the first network data storage location, the processor is configured to:

compute a plurality of cost values for a respective plurality of potential fraction values; and

determine the fraction of content item versions to cache at the first network data storage location based on identification of one of the potential fraction values having a lowest cost value associated therewith.

9. The apparatus of claim 1 , wherein the popularity distribution of the content item versions is a Zipfian (Zipf) distribution.

10. The apparatus of claim 1 , wherein the processor is configured to:

initiate at least one configuration action based on the fraction of content item versions to cache at the first network data storage location.

11. The apparatus of claim 1 , wherein the processor is configured to execute a cache replacement scheme at the first network data storage location based on the fraction of content item versions to cache at the first network data storage location.

12. A non-transitory computer-readable storage medium storing instructions which, when executed by a computer, cause the computer to perform a method, the method comprising:

determining a fraction of a plurality of content item versions to cache at a first network data storage location based on a popularity distribution of the content item versions and based on cost model information associated with the first network data storage location and a second network data storage location, wherein the cost model information comprises at least one of:

a first set of cost model information comprising a cost of storing one of the content item versions at the first network data storage location and a cost of transcoding one of the content item versions at the second network data storage location, wherein the first network data storage location is located at a first hierarchical level and the second network data storage location is located at a second hierarchical level, wherein the first hierarchical level is located farther from end users than the second hierarchical level; or

a second set of cost model information comprising a cost of storing one of the content item versions at the first network data storage location and a cost of transferring one of the content item versions from the second network data storage location to the first network data storage location, wherein the first network data storage location is located at a first hierarchical level and the second network data storage location is located at a second hierarchical level, wherein the first hierarchical level is located closer to end users than the second hierarchical level.

13. A method, comprising:

determining, using a processor, a fraction of a plurality of content item versions to cache at a first network data storage location based on a popularity distribution of the content item versions and based on cost model information associated with the first network data storage location and a second network data storage location, wherein the cost model information comprises at least one of:

a first set of cost model information comprising a cost of storing one of the content item versions at the first network data storage location and a cost of transcoding one of the content item versions at the second network data storage location, wherein the first network data storage location is located at a first hierarchical level and the second network data storage location is located at a second hierarchical level, wherein the first hierarchical level is located farther from end users than the second hierarchical level; or

a second set of cost model information comprising a cost of storing one of the content item versions at the first network data storage location and a cost of transferring one of the content item versions from the second network data storage location to the first network data storage location, wherein the first network data storage location is located at a first hierarchical level and the second network data storage location is located at a second hierarchical level, wherein the first hierarchical level is located closer to end users than the second hierarchical level.

Assignments (10)