Method to increase random I/O performance with low memory overheads

1. A computer-implemented method for storing a data block in a storage system, the method comprising:

in response to a request from a client to store a data block in a storage system, segmenting the data block into a plurality of subblocks;

individually compressing each of the plurality of subblocks into a compressed subblock;

packing the compressed subblocks into a compressed data block;

storing the compressed data block having the individually compressed subblocks therein in a persistent storage device; and

storing metadata of the compressed data block in an index entry in an index of the storage system, including storing subblock locators indicating locations of the compressed subblocks, wherein each of the subblocks can be individually accessed based on a corresponding subblock locator without having to access remaining subblocks.

2. The method of claim 1 , wherein a subblock locator of a subblock includes a start offset of the subblock within the compressed data block.

3. The method of claim 2 , wherein the subblock locator further includes a size of the subblock.

4. The method of claim 1 , wherein the index entry further includes a block start location indicating a storage location of the compressed data block.

5. The method of claim 1 , wherein the index entry further includes an alignment boundary value indicating a storage alignment of a storage device in which the corresponding subblocks are stored.

6. The method of claim 1 , wherein the index entry includes a device identifier identifying one of a plurality of storage devices in which the compressed data block is stored.

7. The method of claim 1 , further comprising, for each of the subblocks, determining whether a size of the subblock is reduced due to data compression, wherein a subblock is compressed only if the size of the subblock is reduced after data compression.

8. A non-transitory machine-readable medium having instructions stored therein, which when executed by a processor, cause the processor to perform operations of storing a data block in a storage system, the operations comprising:

in response to a request from a client to store a data block in a storage system, segmenting the data block into a plurality of subblocks;

individually compressing each of the plurality of subblocks into a compressed subblock;

packing the compressed subblocks into a compressed data block;

storing the compressed data block having the individually compressed subblocks therein in a persistent storage device; and

storing metadata of the compressed data block in an index entry in an index of the storage system, including storing subblock locators indicating locations of the compressed subblocks, wherein each of the subblocks can be individually accessed based on a corresponding subblock locator without having to access remaining subblocks.

9. The machine-readable medium of claim 8 , wherein a subblock locator of a subblock includes a start offset of the subblock within the compressed data block.

10. The machine-readable medium of claim 9 , wherein the subblock locator further includes a size of the subblock.

11. The machine-readable medium of claim 8 , wherein the index entry further includes a block start location indicating a storage location of the compressed data block.

12. The machine-readable medium of claim 8 , wherein the index entry further includes an alignment boundary value indicating a storage alignment of a storage device in which the corresponding subblocks are stored.

13. The machine-readable medium of claim 8 , wherein the index entry includes a device identifier identifying one of a plurality of storage devices in which the compressed data block is stored.

14. The machine-readable medium of claim 8 , wherein the operations further comprise, for each of the subblocks, determining whether a size of the subblock is reduced due to data compression, wherein a subblock is compressed only if the size of the subblock is reduced after data compression.

15. A storage system, comprising:

a processor; and

a memory coupled to the processor to store instructions, which when executed by the processor, cause the processor to perform operations of storing data, the operations including

in response to a request from a client to store a data block in the storage system, segmenting the data block into a plurality of subblocks,

individually compressing each of the plurality of subblocks into a compressed subblock,

packing the compressed subblocks into a compressed data block,

storing the compressed data block having the individually compressed subblocks therein in a persistent storage device, and

storing metadata of the compressed data block in an index entry in an index of the storage system, including storing subblock locators indicating locations of the compressed subblocks, wherein each of the subblocks can be individually accessed based on a corresponding subblock locator without having to access remaining subblocks.

16. The system of claim 15 , wherein a subblock locator of a subblock includes a start offset of the subblock within the compressed data block.

17. The system of claim 16 , wherein the subblock locator further includes a size of the subblock.

18. The system of claim 15 , wherein the index entry further includes a block start location indicating a storage location of the compressed data block.

19. The system of claim 15 , wherein the index entry further includes an alignment boundary value indicating a storage alignment of a storage device in which the corresponding subblocks are stored.

20. The system of claim 15 , wherein the index entry includes a device identifier identifying one of a plurality of storage devices in which the compressed data block is stored.

21. The system of claim 15 , wherein the operations further comprise, for each of the subblocks, determining whether a size of the subblock is reduced due to data compression, wherein a subblock is compressed only if the size of the subblock is reduced after data compression.