SYSTEM AND METHOD OF ACCESSING A DOCUMENT EFFICIENTLY THROUGH MULTI-TIER WEB CACHING

Patent №

US 8,224,964

Granted

2012-07-17

Filed 2004

Owner

GOOGLE INC.

AI components

2

nlp · hardware

Assignment

Recorded

Dataset

AIPD

2023_r1 edition

Application

10882796

A client-server network system implements a multi-tier caching strategy for a user to access a document efficiently. When serving a document to client, the document is examined for embedded references to other documents. The embedded references are extracted and then assigned a priority. The documents referenced are obtained locally from the server cache or obtained from their respective web hosts and transmitted according to the assigned priority. If the client determines that it has the document, the transmission of is terminated. If the client or server determines that a newly requested document is one of the referenced documents being transmitted, the transmission is given a higher priority. In some instances, the server calculates a content difference between a current version of a referenced and a previous version and transmits this content.

Natural languageAI hardwareH04L 67/1097H04L 67/02H04L 67/568H04L 67/60

AI classification

Natural language0.88
AI hardware0.68
Knowledge representation0.17
Machine learning0.01
Planning0.00
Vision0.00
Evolutionary computation0.00
Speech0.00

Ownership

GOOGLE INC.

assignment · 279410157

Assignors

FREDRICKSEN, ERIC RUSSELL, FENG, HANPING, KATARU, NAGA SRIDHAR, HARIK, GEORGES

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

© 2026 NYSGPT2525 LLC