Fault domain technology-based data management method and apparatus

Abstract

The invention relates to a fault domain technology-based data management method. The method comprises the steps of obtaining a file creation operation request, wherein the file creation operation request carries a fault domain level setting parameter; obtaining at least one target domain identifier according to the fault domain level setting parameter; determining a storage domain identifier according to the at least one target domain identifier; and creating a file copy on a host corresponding to the storage domain identifier. According to the method, a user can set a fault domain level of file copy creation according to the need, so that the user can control storage position and range of the file copy creation and the data management is more flexible; and meanwhile, the user controls different file copies to be created in different fault domain levels, so that storage resources of a distributed storage system can be utilized more efficiently and effectively, and the waste of the storage resources is avoided. In addition, the invention provides a fault domain technology-based data management apparatus.

Description

technical field [0001] The invention relates to the field of computer technology, in particular to a data management method and device based on fault domain technology. Background technique [0002] In order to improve the availability of files, a traditional distributed file system generally saves multiple copies of a file in a distributed storage cluster. In this way, when a file is accessed, if one copy cannot be accessed, the other copy can be accessed, thereby improving the availability of the file. In order to ensure the isolation between copies, each file copy is usually stored in a different fault domain, where a fault domain refers to a group of physical infrastructures (including racks, hosts, etc.) divided in the same distributed system. , socket, router, etc.). This group of physical infrastructures will not share a single point of failure with any other physical infrastructure, so that when there is a single point of failure in one fault domain, it will not af...

AI Technical Summary

Problems solved by technology

[0003] In a traditional distributed storage system, one or more racks are usually divided into a single-level fault domain, so that users can control the storage location of file copies only at the rack level, but cannot control the storage location of files. Computer room level or data center level

Obviously, the traditional method is very inflexible, which is not conducive to the rational utilization of storage resources

Images