Method for a storage device processing data and storage device
A method for managing data in a hierarchical storage device which includes primary storage disks and secondary storage disks. The primary storage disks have higher performance than the secondary storage disks. The storage device detects an access frequency of a data block stored in the secondary storage disks. Based on the access frequency of a data block, the storage device determines that the access frequency of the data block reaches an access threshold value. And then, based upon the determination, the storage device moves the data block to the primary storage disks. After that, the storage device divides the data block into a plurality of sub-blocks and detects an access frequency of each of the sub-blocks. Finally, the storage device moves one or more of the sub-blocks of which access frequencies are less than the access threshold value back to the secondary storage disks.
1. A method for managing data in a hierarchical storage device which includes primary storage disks and secondary storage disks, wherein the primary storage disks have higher performance than the secondary storage disks, the method performed by the storage device comprising:
detecting an access frequency of a data block stored in the secondary storage disks;
determining that the access frequency of the data block reaches an access threshold value;
moving the data block to the primary storage disks based upon determining that the access frequency of the data block reaches the access threshold value;
dividing the data block in the primary storage disks into a plurality of sub-blocks, wherein the sub-blocks are independently accessible, and wherein each of the sub-blocks has a respective access frequency separate from access frequencies of other sub-blocks;
detecting the access frequencies of the sub-blocks; and
moving a sub-block back to the secondary storage disks when an access frequency of the sub-block is less than the access threshold value.
2. The method according to claim 1 , further comprising recording a relationship between the data block and each of the sub-blocks.
3. A hierarchical storage device comprising:
primary storage disks for storing data;
secondary storage disks for storing data, wherein the primary storage disks have higher performance than the secondary storage disks; and
a processor connected to the primary storage disks and the secondary storage disks, and configured to:
detect an access frequency of a data block stored in the secondary storage disks;
determine that the access frequency of the data block reaches an access threshold value;
move the data block to the primary storage disks based upon determining that he access frequency of the data block reaches the access threshold value;
divide the data block in the primary storage disks into a plurality of sub-blocks, wherein the sub-blocks are independently accessible, and wherein each of the sub-blocks has a respective access frequency separate from access frequencies of other sub-blocks;
detect the access frequencies of the sub-blocks; and
move a sub-block back to the secondary storage disks when an access frequency of the sub-block is less than the access threshold value.
4. The device according to claim 3 , wherein the processor is further configured to record a relationship between the data block and each of the sub-blocks.