EFFICIENT ONE-TO-MANY CONTENT DISTRIBUTION IN A PEER-TO-PEER COMPUTER NETWORK

Patent №

US 7,593,333

Granted

2009-09-22

Filed 2004

Owner

MICROSOFT CORPORATION

AI components

1

kr

Assignment

Recorded

Dataset

AIPD

2023_r1 edition

Application

10887406

A content distribution method and system for distributing content over a peer-to-peer network such that the full potential throughput of the network is achieved. The content distribution method divides the content to be distributed into many small blocks. Each of the content blocks then is assigned to a node, which can be a content-requesting node, a non-content-requesting node or a source node. Content is assigned based on a capacity of the node, where nodes having a larger capacity are assigned a greater number of content blocks and nodes having a smaller capacity are assigned a fewer content blocks. The capacity generally is defined as the upload bandwidth of the node. Redistribution queues are employed to control the throughput of the distribution. This bandwidth control strategy ensures that upload bandwidths of the peer and source nodes are fully utilized even with network anomalies such as packet losses and delivery jitters.

Knowledge representationH04L 12/1877H04L 12/1854H04L 65/70H04L 67/06H04L 67/104H04L 67/1076H04L 67/108H04L 69/16+3 more

AI classification

Knowledge representation0.55
Machine learning0.01
AI hardware0.00
Evolutionary computation0.00
Natural language0.00
Planning0.00
Vision0.00
Speech0.00

Ownership

MICROSOFT CORPORATION

assignment · 155630645

Assignors

LI, JIN, CHOU, PHILIP A., ZHANG, CHA

On an employer assignment, the assignors are typically the inventors.

© 2026 NYSGPT2525 LLC