Method for Synchronizing Software Application and User Data for Asynchronous Client-Server and Peer to Peer Computer Networks

• US 20070276836A1
• Filed: 05/08/2006
• Published: 11/29/2007
• Est. Priority Date: 05/08/2006
• Status: Active Grant

• Current Assignee: Qualcomm, Inc.
• Original Assignee: Gregory Simon, Manjirnath Chatterjee
• Inventors: Chatterjee, Manjirnath, Simon, Gregory
• Granted Patent: US 7,860,825 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/182 Distributed file systems

Abstract

Web Applications run as software which is served through a portal or browser. However this precludes their use offline or when connectivity is not present. The present invention is a method for creating offline web applications and managing associated user data which is created offline and must later be reconciled with one or more central servers without a data collision. This allows the system to efficiently know which version of different pieces of user data are new and which need to be added to centralized servers. This method also applies to the actual web application program files so that software applications can be synchronized in addition to user data enabling a transparent online and offline user experience. The method described here also applies to mobile computing devices such as cellular telephones and personal digital assistants. A side benefit of the method introduced here is that data sets can be distributed in manner which allows peer to peer synchronization and filedata distribution. Lastly the method described herein minimizes the amount of transactional data required to synchronize data sets across a network thereby increasing efficiency of available bandwidth on a computer network using the system.

Claims

1. A method for creating a distributed data sharing system for storing and synchronizing data in a computer network where:

a. Data is stored and referenced as Sets and Items within a Set

b. Universally Unique Identifiers are used as the means for identifying Sets and the Items within a Set at any node in the network

c. where the Item IDs are created via a norm mechanism which uses item data to generate IDs from sufficiently large one way hash function

d. where the Items have separated metadata and payload sections.

2. A method of claim 1 where the and the only item level operators are add and delete.

3. A method of storage of data where the items of claim 1 are stored as rows in a table and the metadata portions, and payload portions are columns in a table such as in a database.

4. A method of claim 1 where a normalizing sequence for metadata for hash generation which is used for the global ID number of a data item.

5. A method of claim 1 where the item metadata can be stored as binary information, textual information or XML.

6. A Method for synchronizing the data in claim 1 where the synchronization occurs only in 1 direction as in a master—

slave system.

7. A Method of claim 1 where some item files act as Manifest files which describe which items belong to a chunked piece of data which is spread over multiple items within a set for the purpose of peer to peer use as a distributed file system.

8. A method for efficient set difference determination by using the unique.

9. A method of claim 8 where the use of system specific rules to prioritize which IDs to synchronize so as to conserve bandwidth for the determination of which items to synchronize.

10. A method of claim 8 for segmentation of IDs via ranges to make the determination of which items to synchronize efficient.

11. A method of claim 8 where the process of determining which items to sync is differential in nature and based on a list of the previously synchronized items.