A light center heavy agent multi-cloud data unified management and control method and system

By deploying lightweight data acquisition agents and serverless execution functions in a multi-cloud environment, decoupling and self-healing of cross-cloud data management and control are achieved, solving the bandwidth bottleneck and consistency issues of centralized gateways and improving the reliability and economy of data governance.

CN122372631APending Publication Date: 2026-07-10XINJIANG YUNRUI INFORMATION TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN Β· China
Patent Type
Applications(China)
Current Assignee / Owner
XINJIANG YUNRUI INFORMATION TECH CO LTD
Filing Date
2026-05-24
Publication Date
2026-07-10

AI Technical Summary

Technical Problem

In existing technologies, the centralized gateway architecture for cross-cloud data management becomes a bottleneck in the face of petabyte-level massive data or high-concurrency transmission scenarios, as the network bandwidth and processing capacity of the central node become bottlenecks, and the risk of single point of failure is high. Traditional solutions cause great interference to source-end services and are difficult to solve the problem of data consistency within the transmission window.

Method used

A unified multi-cloud data management approach with a lightweight central processing unit and a heavy agent system is adopted. By deploying a lightweight collection agent, metadata is scanned at the heterogeneous storage source and synchronized to the central management node. Serverless execution functions are used to directly perform data operations between the source storage address and the target storage address. Combined with distributed message services and edge handshake atomicity verification mechanisms, data scheduling and transmission are decoupled and self-healed.

Benefits of technology

It breaks through the bandwidth bottleneck of traditional centralized gateways, realizes horizontal linear expansion of data throughput capacity, reduces the impact on source-end production business, solves the data consistency problem in wide area network environment, and improves the reliability and economy of data governance in multi-cloud environment.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122372631A_ABST
    Figure CN122372631A_ABST
Patent Text Reader

Abstract

This invention provides a method and system for unified management and control of multi-cloud data, characterized by a lightweight central control and a heavy agent-based approach. The method utilizes lightweight acquisition agents deployed at various heterogeneous storage sources in multiple clouds to scan and extract metadata information, synchronizing this metadata to a central management node to construct a global data directory. The central management node performs intelligent analysis based on the global data directory and preset management strategies, generating data scheduling instructions and publishing them to a distributed messaging service. A pre-built serverless execution function in the execution environment subscribes to and consumes these instructions, directly executing data entity operations between the source and target storage addresses in response. After completing the data entity operations, the serverless execution function returns the execution result to the central management node via a callback interface. This invention solves the bandwidth bottleneck problem of the central node and utilizes edge handshake and backpressure mechanisms to ensure data consistency and source-end business stability.
Need to check novelty before this filing date? Find Prior Art