Method and system for allocating content using a content-based centrality metric
AssigneeFuturewei Technologies, Inc.
InventorCedric Westphal
Life of this patent
Estimating the term…
Abstract
The present technology provides a new content-based centrality metric that may be used for allocating content to caches in a network. The proposed method for measuring centrality and allocating content not only considers the topology of the network, but also the distribution of cache spaces and content among the cache spaces in the network in order to improve how content may be allocated. This approach to content allocation enables improved content retrieval in the form of higher cache hit rates, lower delays and improved network performance.
Claims22
A network content allocation device for allocating content items in a network comprising a plurality of nodes and one or more user devices, the network content allocation device comprising:
- a non-transitory memory storage comprising instructions;
- and
- one or more processors in communication with the memory, wherein the one or more processors execute the instructions to:
- determine a content-based centrality score for each of the plurality of nodes, the content based centrality score for a node determined by computing a centrality of the node based on a location of the node relative to locations of the one or more user devices and relative to the content items in the network, and
- allocate content items to the plurality of nodes based on a comparison of content-based centrality scores among nodes in the network.
A computer-implemented method for allocating one or more content items among one or more caches in a network, comprising:
- determining a content-based centrality score for each of the plurality of nodes, the content based centrality score for a node determined by computing a centrality of the node based on a location of the node relative to locations of the one or more user devices and relative to the content items in the network, and
- allocating content items to the plurality of nodes based on a comparison of content-based centrality scores among nodes in the network.
The computer-implemented method of claim 17, wherein the memory of the plurality of nodes comprises a common cache space storing content items that are stored in each of the plurality of nodes, and a specific cache space storing content items that are stored in less than each of the plurality of nodes.
A network content allocation device for allocating content items in a network comprising a plurality of nodes and one or more user devices, the network content allocation device comprising:
- a non-transitory memory storage comprising instructions;
- and
- one or more processors in communication with the memory, wherein the one or more processors execute the instructions to:
- determine a popularity-weighted, content-based centrality score for each of the plurality of nodes, the content based centrality score for a node determined by computing a weighted popularity of content items in the node, and a centrality of the node based on a location of the node relative to locations of the one or more user devices and relative to the content items in the network, and
- allocate content items to the plurality of nodes based on a comparison of popularity-weighted, content-based centrality scores among nodes in the network.
A network content allocation device for allocating content items in a network comprising a plurality of nodes and one or more user devices, the plurality of nodes comprising a plurality of caches, the network content allocation device comprising:
- a non-transitory memory storage comprising instructions;
- and
- one or more processors in communication with the memory, wherein the one or more processors execute the instructions to:
- determine a popularity-weighted, content-based centrality score for each of the plurality of nodes, the popularity-weighted, content based centrality score for a node determined using a pre-computed centrality measure, the pre-computed centrality measure computed by:
- determining a number of different combinations of the plurality of caches,
- determining, for the plurality of different combinations of caches, the pre-computed centrality measure for a single content item based on a location of the plurality of caches relative to locations of the one or more user devices relative to the single content item and using a predefined popularity value for the single content item, and
- storing the pre-computed centrality measure for the single content item in the plurality of caches.
Dependencies are inferred from the claim text.