Storage unit allocation methods, devices, servers, storage systems, and media

CN115586862BActive Publication Date: 2026-05-26BEIJING KINGSOFT CLOUD NETWORK TECH CO LTD
View PDF 1 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
BEIJING KINGSOFT CLOUD NETWORK TECH CO LTD
Filing Date
2021-07-05
Publication Date
2026-05-26

AI Technical Summary

Technical Problem

In a distributed storage system, the process of the data server requesting storage unit identifiers from the meta server and the meta server allocating storage unit identifiers takes a long time, resulting in low data writing efficiency and the possibility of write failures.

Method used

The data server periodically checks the number of allocated storage units on the target disk. When the number is less than a preset threshold, it generates an allocation request to request storage unit identifiers from the meta server, thereby pre-allocating storage unit identifiers and reducing the reliance on the meta server during the writing process.

Benefits of technology

By pre-allocating storage unit identifiers, data write time is reduced, write failures are avoided, and data write efficiency is improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115586862B_ABST
    Figure CN115586862B_ABST
Patent Text Reader

Abstract

This disclosure provides a method, apparatus, server, storage system, and medium for allocating storage units. When the number of first storage units on a target disk is less than a preset threshold, an allocation request is generated. The first storage unit refers to a storage unit that has been allocated a storage unit identifier and is not yet full of data. The allocation request is used to request a metaserver to allocate storage unit identifiers to some or all of the second storage units on the target disk. The second storage units refer to storage units that have not been allocated storage unit identifiers. The allocation request is then sent to the metaserver. This achieves pre-allocation of storage unit identifiers. During data writing, the pre-allocated storage unit identifiers can be directly obtained, reducing the time consumed by data writing and improving data writing efficiency.
Need to check novelty before this filing date? Find Prior Art