Non-volatile cache in disk drive emulation

• US 8,024,515 B2
• Filed: 12/07/2007
• Issued: 09/20/2011
• Est. Priority Date: 12/07/2007
• Status: Expired due to Fees

• Current Assignee: Western Digital Technologies Incorporated (Western Digital Corporation)
• Original Assignee: Hitachi Global Storage Technologies Netherlands BV (Western Digital Corporation)
• Inventors: Auerbach, Daniel J., Ng, Spencer W.
• Primary Examiner(s): Bataille; Pierre-Michel
• Assistant Examiner(s): Fishburn; John P
• Application Number:
  US11/952,534
• Publication Number:
  US 20090150614A1
• Time in Patent Office: 1,383 Days
• Field of Search: None
• US Class Current: 711/113
• CPC Class Codes: G06F 12/0804 with main memory updating G06F12/0806 takes precedence
• CPC Class Codes: G06F 12/0866 for peripheral storage systems, e.g. disk cache
• CPC Class Codes: G06F 2212/221 Static RAM
• CPC Class Codes: G06F 2212/222 Non-volatile memory
• CPC Class Codes: G06F 2212/461 Sector or disk block
• CPC Class Codes: G06F 3/061 Improving I/O performance
• CPC Class Codes: G06F 3/0625 Power saving in storage systems
• CPC Class Codes: G06F 3/064 Management of blocks
• CPC Class Codes: G06F 3/0656 Data buffering arrangements
• CPC Class Codes: G06F 3/068 Hybrid storage device
• CPC Class Codes: Y02D 10/00 Energy efficient computing, e.g. low power processors, power management or thermal management

Abstract

A method and apparatus for deferring media writes for emulation drives are provided. By deferring media writes using non-volatile storage, the performance penalty associated with RMW operations may be minimized. Deferring writes may allow the RMW operations to be done while the disk drive is idle. Further, deferring writes may also allow data blocks to be accumulated over time, allowing a full (4K) disk drive block size to be written with a simple write operation, thus making a RMW unnecessary.

Claims

1. A method for disk drive emulation, comprising:

receiving a request to write data to a storage medium of a hard drive, wherein the request contains write data comprising one or more blocks formatted based on a first block size while the storage medium comprises a plurality of blocks formatted based on a second block size larger than the first block size;

caching the write data in non-volatile memory on the hard drive;

upon determining that the write data does not align with a start or an end of a block in the storage medium, deferring writing the write data to the storage medium;

combining the write data with different write data received via additional write data requests; and

upon detecting that the write data aligns with both a start and an end of a block in the storage medium, writing the write data from the non-volatile memory to the storage medium.

2. The method of claim 1, wherein writing the write data from the non-volatile memory to the hard drive storage medium comprises, performing a read-modify-write by:

reading a second block size of data from the storage medium;

modifying the data read from the storage medium to include the write data; and

writing the modified data back to the storage medium.

3. The method of claim 1, further comprising:

maintaining the write data in the non volatile storage after writing the write data to the storage medium; and

marking the space occupied by the write data as free.

4. The method of claim 1, wherein the hard drive storage medium is formatted based on a hard drive block size of 4 KB.

5. The method of claim 1, comprising deferring writing the write data to the storage medium across one or more power cycles.

6. The method of claim 5, further comprising:

detecting that there is write data in the non volatile memory after a power on reset; and

in response, writing the write data to the storage medium without receiving another write request to write the write data to the storage medium.

7. A hard disk drive (HDD) controller configured to:

receive a request to write data to a storage medium of a hard drive, wherein the request contains write data comprising one or more blocks formatted based on a first block size while the storage medium comprises a plurality of blocks formatted based on a second block size larger than the first block size;

cache the write data in non-volatile memory on the hard drive;

upon determining that the write data does not align with a start or an end of a block in the storage medium, defer writing the write data to the storage medium;

combine the write data with different write data received via additional write data requests; and

upon detecting that the write data aligns with both a start and an end of a block in the storage medium, write the write data from the non-volatile memory to the storage medium.

8. The HDD controller of claim 7, wherein the controller is configured to write the write data from the non-volatile memory to the hard drive storage medium, wherein the write comprises, performing a read-modify-write by:

reading a second block size of data from the storage medium;

modifying the data read from the storage medium to include the write data; and

writing the modified data back to the storage medium.

9. The HDD controller of claim 7, wherein the controller is further configured to:

maintain the write data in the non volatile storage after writing the write data to the storage medium; and

mark the space occupied by the write data as free.

10. The HDD controller of claim 7, wherein the hard drive storage medium is formatted based on a hard drive block size of 4 KB.

11. The HDD controller of claim 7, wherein the HDD controller may defer writing the write data to the storage medium across one or more power cycles.

12. The HDD controller of claim 11, wherein the controller is further configured to:

detect if there is write data in the non volatile memory after a power on reset; and

in response, write the write data to the storage medium without receiving another write request to write the write data to the storage medium.

13. A hard disk drive, comprising:

at least one long-term storage medium;

one or more segments of non-volatile memory; and

a controller configured to;

receive a request to write data to a storage medium of a hard drive, wherein the request contains write data comprising one or more blocks formatted based on a first block size while the storage medium comprises a plurality of blocks formatted based on a second block size larger than the first block size;

cache the write data in non-volatile memory on the hard drive;

upon determining that the write data does not align with a start or an end of a block in the storage medium, defer writing the write data to the storage medium;

combine the write data with different write data received via additional write data requests; and

upon detecting that the write data aligns with both a start and an end of a block in the storage medium, write the write data from the non-volatile memory to the storage medium.

14. The hard disk drive of claim 13, wherein the controller is further configured to:

maintain the write data in the non volatile storage after writing the write data to the storage medium; and

mark the space occupied by the write data as free.

15. The hard disk drive of claim 13, wherein the hard drive storage medium is formatted based on a hard drive block size of 4 KB.

16. The hard disk drive of claim 13, wherein the HDD controller may defer performance of write operations over a power cycle.

17. The hard drive disk of claim 13, wherein the controller is further configured to:

detect if there is write data in the non volatile memory after a power on reset; and

in response, write the write data to the storage medium without receiving another write request to write the write data to the storage medium.