Method and Distributed Computer System for Synchronizing Data Sets Stored in Different Communication Devices

Simple SummaryContent extracted from patent full text and abstract with AI.

This invention describes a method and a distributed computing system for synchronizing data sets stored on different communication devices, such as computers, smartphones, or sensors, even in cases where the devices are only intermittently connected. The system utilizes random linear codes—commonly used for error correction—not only to ensure reliable data transfer over unreliable communication links but also to reduce the computational and communication overhead during synchronization. By partitioning large data sets into smaller subsets and synchronizing these subsets individually, the approach makes the synchronization process more scalable, efficient, and robust against data loss.

Use CasesContent extracted from patent full text and abstract with AI.

• Synchronizing contact lists or files between smartphones and computers over Bluetooth or Wi-Fi where the connection may be unstable.
• Data synchronization in distributed sensor networks where energy and bandwidth are limited.
• Cloud-to-device or device-to-device data backup and restoration, especially over lossy networks.
• Collaborative editing systems where multiple users update shared documents from different devices.
• Incremental synchronization of large datasets in enterprise or IoT environments where only subsets of data change.
• Database replication across geographically distributed data centers or edge devices.

BenefitsContent extracted from patent full text and abstract with AI.

• Significantly reduces the amount of data that needs to be transferred during synchronization, saving bandwidth.
• Decreases computational complexity and energy consumption, especially on battery-powered devices.
• Scalable to very large datasets by breaking the synchronization problem into smaller, more manageable subproblems.
• Provides robustness against data loss during transmission, thanks to integrated error correction.
• Supports bidirectional and flexible synchronization strategies based on the specific needs of connected devices.
• Enables efficient synchronization even over error-prone or intermittent communication channels.
• Limits communication overhead to the number of actual differences between datasets, rather than the total dataset size.

Technical Classifications (CPCs)

Inventors & Applicants

Patent Abstract

A method and a distributed computing system (1) are provided to use a forward error correction code for both data synchronization as well as an error-resistant data transmission during the data synchronization process. In particular, a random linear code is used which is traditionally used as an error correction code. In order to synchronize large data-sets by using random linear codes a set partitioning is performed on the large data sets to divide each data-set into a plurality of data-subsets. Therefore, the distributed computing system (1) is made scalable by dividing the data synchronization issue into many smaller problems. This results in significantly lower computational complexity at the decoding communication device (20), allowing the data synchronization to complete much faster (lesser computation time of the system) and also save communication between the two synchronizing communication devices (10, 20).