SYSTEM AND METHOD OF MANAGING METADATA

• US 20100070729A1
• Filed: 09/18/2008
• Published: 03/18/2010
• Est. Priority Date: 09/18/2008
• Status: Active Grant

• Current Assignee: Seagate Technology LLC (Seagate Technology Holdings Plc)
• Original Assignee: Seagate Technology LLC (Seagate Technology Holdings Plc)
• Inventors: Ng, Wei Loon, Shen, Feng, Stefanus, Stefanus
• Granted Patent: US 8,762,661 B2
• Time in Patent Office: Days
• US Class Current: 711/166
• CPC Class Codes: G06F 3/0617 in relation to availability
• CPC Class Codes: G06F 3/0631 by allocating resources to storage systems
• CPC Class Codes: G06F 3/0641 De-duplication techniques
• CPC Class Codes: G06F 3/0656 Data buffering arrangements
• CPC Class Codes: G06F 3/0679 Non-volatile semiconductor memory device, e.g. flash memory, one time programmable memory [OTP]

Abstract

In a particular embodiment, a controller is adapted to control read/write access to a storage media including a pre-allocated area having multiple meta-blocks. The controller includes logic adapted to control the multiple meta-blocks as a first in first out (FIFO) circular queue. The logic selects one or more meta-blocks from the multiple meta-blocks based on an order associated with the FIFO circular queue and selectively writes a logical block address (LBA) mapping table to the selected one or more meta-blocks.

Claims

1. A controller adapted to control read/write access to a storage media including a pre-allocated area having multiple meta-blocks, the controller including logic adapted to control the multiple meta-blocks as a first in first out (FIFO) circular queue, the logic to select one or more meta-blocks from the multiple meta-blocks based on an order associated with the FIFO circular queue and to selectively write a logical block address (LBA) mapping table to the selected one or more meta-blocks.

2. The controller of claim 1, wherein the controller includes logic to store a table flag with the LBA mapping table at the selected one or more meta-blocks.

3. The controller of claim 1, wherein the controller includes logic to load the LBA mapping table from the multiple meta-blocks upon restart.

4. The controller of claim 3, wherein the controller includes logic to determine whether more than one LBA mapping table is stored at the multiple meta-blocks and to select a LBA mapping table based on a table flag and a location of the selected LBA mapping table within the multiple meta-blocks.

5. The controller of claim 4, wherein the logic is adapted to verify a next free erased block and a next block waiting to be erased identified by the selected LBA mapping table.

6. The controller of claim 5, wherein the logic is adapted to read spare data from the next free erased block and from the next block waiting to be erased to determine whether the selected LBA mapping table is correct and to update the selected LBA mapping table when the selected LBA mapping table is incorrect.

7. The controller of claim 1, wherein the controller is adapted to update the LBA mapping table after writing data to the storage media, after a garbage collection operation, and after a pinned set of data is removed.

8. A method comprising:

writing data to a storage media of a storage device;

concurrently writing spare data of a first sector or page with a mapped logical block address, a next free erased block, and a next free block waiting to be erased; and

storing a mapping table and an associated flag at a pre-allocated portion of the storage media at a next meta-block of a plurality of meta-blocks configured to operate as a circular buffer.

9. The method of claim 8, further comprising:

restarting and reinitializing the storage device;

selecting a most recent mapping table from the pre-allocated portion of the storage media based on a flag associated with the mapping table and based on a location of the mapping table within the plurality of meta-blocks.

10. The method of claim 9, further comprising:

locating a next free erased block at the storage media;

reading spare data from the next free erased block to determine whether the next free erased block is written; and

updating the selected mapping table when the next free erased block is written.

11. The method of claim 9, further comprising:

locating a next block to be erased at the storage media;

reading spare data from the next block to be erased to determine whether the next block to be erased is already erased; and

updating the selected mapping table when the next block to be erased is already erased.

12. The method of claim 8, further comprising:

selecting a mapping table from the pre-allocated area of the solid-state storage media based on an associated flag and a location of the selected mapping table within the pre-allocated area;

loading the selected mapping table;

identifying one of a next free erased block and a next block waiting to be erased;

reading spare data associated with the identified one of the next free erased block and the next block waiting to be erased; and

updating the selected mapping table when the spare data indicates that the selected mapping table is out of date.

13. The method of claim 8, wherein the mapping table is written to the pre-allocated area in a first in first out (FIFO) order.

14. The method of claim 8, further comprising:

selecting a mapping table from the pre-allocated area of the solid-state storage media based on an associated flag and a location of the selected mapping table within the pre-allocated area;

loading the selected mapping table;

identifying one of a next free erased block and a next block waiting to be erased;

reading spare data associated with the identified one of the next free erased block and the next block waiting to be erased to detect data corruption; and

updating the selected mapping table in response to detection of data corruption.

15. A system comprising:

a storage media adapted to store data, the storage media comprising a pre-allocated area including multiple meta-blocks to store one or more mapping tables; and

a controller adapted to control read/write access to the storage media, the controller including logic adapted to store a mapping table at one or more selected meta-blocks of the pre-allocated area as a first in first out (FIFO) circular queue, the logic to determine whether more than one mapping table is stored at the multiple meta-blocks and to select the mapping table based on a table flag and a location of the selected mapping table within the multiple meta-blocks upon restart and re-initialization after a power loss.

16. The storage device of claim 15, wherein the controller includes logic to select one or more meta-blocks from the multiple meta-blocks based on an order associated with the FIFO circular queue and to selectively write a logical block address (LBA) mapping table to the selected one or more meta-block.

17. The storage device of claim 15, wherein the controller further includes logic adapted to verify at least one of a next free erased block and a next block waiting to be erased identified by the selected mapping table.

18. The storage device of claim 17, wherein the logic is adapted to examine spare data stored at one of the next free erased block and the next block waiting to be erased to detect data corruption.

19. The storage device of claim 18, wherein the logic is adapted to reconstruct the selected mapping table from the spare data when data corruption is detected.

20. The storage device of claim 15, further comprising:

an interface responsive to a host system and adapted to communicate data between the controller and the host system;

a rotatable storage media adapted to store user data, wherein the selected mapping table is related to metadata stored at a solid-state memory device and to user data stored at the rotatable storage media;

an actuator assembly coupled to the controller and adapted to position a read/write head over the rotatable storage media;

a spindle motor adapted coupled to the controller and adapted to rotate the rotatable storage media; and

firmware executable by the controller to identify a most recent mapping table from one or more mapping tables stored at the pre-allocated area of the storage media and to verify the most recent mapping table against spare data stored in a first page or sector of a particular data block identified by the most recent mapping table.