| 摘要 |
A method and apparatus for managing loads of multiple storage disks, a file system, and a storage network system. The method comprises: firstly, acquiring a storage disk list comprising an identification mark of each storage disk; then, after a file access request is received, extracting file full-path information in the file access request; and selecting, according to the file full-path information and the identification mark of each storage disk, one of the storage disks to serve as a target storage disk accessed according to the file access request by using a hash algorithm. That is, a multi-disk load balance mechanism is implemented by using a hash algorithm, massive files can be quite evenly distributed on multiple disks, existence of metadata is not needed, a system structure becomes very simple and efficient, the requirement for hardware (mainly referring to a memory) is low, a single-point fault caused by damage to the metadata does not exist due to no metadata, and the security of system storage can be improved. |
