Invention Title:

NETWORK-BASED ERASURE-CODED MEMORY RESERVE

Publication number:

US20260172472

Publication date:
Section:

Electricity

Class:

H04L67/1097

Inventors:

Assignee:

Applicant:

Smart overview of the Invention

A network-based system is designed to manage memory across multiple devices using erasure-coded memory reserves. Each device within a network is configured with an erasure-coded memory that stores a portion, or slice, of common data. These slices collectively form a memory reserve that allows for the reconstruction of the full data set without relying on a central data source. This decentralized approach improves data access and reliability across the network.

Background

In traditional network services, devices access common data from a central system, leading to increased data duplication and network load. Each device creates a local copy of the data, which can be inefficient and resource-intensive. The current invention addresses these issues by distributing data slices across devices, reducing the need for full data copies and minimizing demands on central resources.

Implementation

Devices are configured to store erasure-coded slices of common data, enabling them to collaborate and maintain data integrity. When a device requires data not locally stored, it requests the necessary slices from other devices. This method allows for data reconstruction without needing to access a central source, enhancing data availability and reducing network traffic.

Technical Details

Erasure coding involves splitting data into multiple slices with additional redundant information for fault tolerance. This technique ensures that even if some slices are lost, the original data can be reconstructed from the remaining slices. The system supports both single-piece and segmented data storage, allowing for flexible data management across various devices.

Benefits and Updates

The system facilitates efficient data updates by propagating changes through the erasure-coded memory reserve. This reduces the number of requests needed for data updates, preventing network congestion and overloading of individual devices. The distributed approach ensures that no single device becomes a bottleneck, promoting a balanced and resilient network architecture.