MANAGING CACHE DATA AND METADATA

• US 20100070701A1
• Filed: 11/14/2008
• Published: 03/18/2010
• Est. Priority Date: 09/15/2008
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Bak, Yevgeniy, Ergan, Cenk, Kirshenbaum, Alexander, Iyigun, Mehmet, Fortin, Michael, Fields, David
• Granted Patent: US 9,032,151 B2
• Time in Patent Office: Days
• US Class Current: 711/113
• CPC Class Codes: G06F 11/1441 Resetting or repowering
• CPC Class Codes: G06F 12/0833 in combination with broadcast means (e.g. for invalidation or updating)
• CPC Class Codes: G06F 12/0862 with prefetch
• CPC Class Codes: G06F 12/0866 for peripheral storage systems, e.g. disk cache
• CPC Class Codes: G06F 12/128 adapted to multidimensional cache systems, e.g. set-associative, multicache, multiset or multilevel
• CPC Class Codes: G06F 12/14 Protection against unauthorised use of memory or access to memory
• CPC Class Codes: G06F 2212/1052 Security improvement
• CPC Class Codes: G06F 2212/2022 Flash memory
• CPC Class Codes: G06F 2212/2146 being detachable, e.g.. USB memory
• CPC Class Codes: G06F 2212/222 Non-volatile memory
• CPC Class Codes: G06F 2212/466 Metadata, control data
• CPC Class Codes: G06F 2212/62 Details of cache specific to multiprocessor cache arrangements
• CPC Class Codes: Y02D 10/00 Energy efficient computing, e.g. low power processors, power management or thermal management

Abstract

Embodiments of the invention provide techniques for ensuring that the contents of a non-volatile memory device may be relied upon as accurately reflecting data stored on disk storage across a power transition such as a reboot. For example, some embodiments of the invention provide techniques for determining whether the cache contents and/or or disk contents are modified during a power transition, causing cache contents to no longer accurately reflect data stored in disk storage. Further, some embodiments provide techniques for managing cache metadata during normal (“steady state”) operations and across power transitions, ensuring that cache metadata may be efficiently accessed and reliably saved and restored across power transitions.

Claims

1. A method for operating a computer (700) having coupled thereto a storage medium (706) and a cache device (110), the computer comprising an operating system, the method comprising acts of:

(A) in response to a request to write a data item to an address on the storage medium, causing the data item to be written to the address on the storage medium (101) and to a corresponding address on the cache device (102);

(B) subsequent to a reboot of the operating system, determining (270, 330) whether the data item stored at the address on the cache device can reliably be employed to satisfy a request to read the data item from the address on the storage medium; and

(C) if it is determined in (B) that the data item stored at the address on the cache device can reliably be employed to satisfy a request to read the data item from the address on the storage medium, causing (299) the data item to be read from the address on the cache device.

2. The method of claim 1, wherein the acts (A)-(C) are performed by the operating system.

3. The method of claim 1, wherein the storage medium comprises a disk storage device.

4. The method of claim 1, wherein the cache device is removable from the computer.

5. The method of claim 1, wherein the act (B) is performed in response to a request to read the data item from the address on the storage medium, and further comprises verifying that the data item stored at the address on the cache device was not modified after completion of the act (A).

6. The method of claim 5, wherein (A) further comprises generating a representation of at least the data item and writing the representation to the cache device, and wherein the act of verifying in (B) comprises:

(B1) retrieving the representation written to the cache device;

(B2) re-generating the representation; and

(B3) comparing the representation retrieved in (B1) to the representation re-generated in (B2) to determine whether the data item stored at the address on the cache device can reliably be employed to satisfy the request to read the data item.

7. The method of claim 1, further comprising an act of providing a write recorder component operable to identify write operations performed to the storage medium which are not performed to the cache device, and wherein the act (B) further comprises determining, via the write recorder component, whether the data item stored at the address on the storage medium was modified after completion of the act (A).

8. At least one computer storage medium (706) having stored thereon instructions which, when executed by a computer comprising an operating system and having coupled thereto a cache device (110), perform a method comprising acts of:

(A) in response to a request to write a data item to an address on the computer storage medium, causing the data item to be written to the address on the computer storage medium (101) and to a corresponding address on the cache device (102);

(B) subsequent to a reboot of the operating system and in response to a request to read the data item from the address on the computer storage medium, determining (270) whether the data item stored at the address on the cache device was modified after (A); and

(C) if it is determined in (B) that the data item stored at the address on the cache device was not modified after (A), causing (299) the data item to be read from the address on the cache device.

9. The at least one computer storage medium of claim 8, further comprising an act of:

(D) if it is determined in (B) that the data item stored at the address on the cache device was modified after (A), causing the data item to be read from the address on the computer storage medium.

10. The at least one computer storage medium of claim 9, wherein the act (D) further comprises causing the data item to be evicted from the address on the cache device.

11. The at least one computer storage medium of claim 8, wherein the act (A) further comprises generating a representation of at least the data item and writing the representation to the cache device, and wherein the act of determining in (B) comprises:

(B1) retrieving the representation written to the cache device;

(B2) re-generating the representation; and

(B3) comparing the representation retrieved in (B1) to the representation re-generated in (B2) to determine whether the data item stored at the address on the cache device can reliably be employed to satisfy the request to read the data item.

12. The at least one computer storage medium of claim 8, wherein the act (A) further comprises generating a representation of at least the data item and writing the representation to the cache device, and wherein the act of generating comprises employing a cryptographic technique.

13. The at least one computer storage medium of claim 8, wherein the act (A) further comprises employing cache metadata to determine the address on the cache device to which the data item is written, and wherein the act (B) further comprises determining whether the cache metadata was modified after (A).

14. A computer (700), comprising:

a storage medium (706) having at least one address at which data is stored;

a cache device (110) having at least one address at which data is stored;

at least one processor (703) programmed via an operating system to;

in response to a request to write a data item to a first address on the storage medium, cause the data item to be written to the first address on the storage medium (101) and to a corresponding address on the cache device (102);

identify (300) an occurrence of a write operation to an address on the storage medium which is not also performed to an address on the cache device; and

determine (30), subsequent to a reboot of the operating system, whether the data item stored at the corresponding address on the cache device can reliably be used to satisfy a request to read the data item from the first address on the storage medium, by determining whether a write operation was performed to the first address on the storage medium which was not also performed to the corresponding address on the cache device.

15. The computer of claim 14, wherein the at least one processor is programmed to identify an occurrence of a write operation to an address on the storage medium which is not also performed to an address on the cache device using a log which is written to the storage medium and which is accessed upon reboot of the operating system.

16. The computer of claim 14, wherein the at least one processor is programmed to, if it is determined that the data item stored at the corresponding address on the cache device can not reliably be used to satisfy a request to read the data item from the first address on the storage medium, evict the data item from the corresponding address on the cache device and satisfy the request by reading the data item from the first address on the storage medium.

17. The computer of claim 14, wherein the identification of a write operation to an address on the storage medium that is not also performed to an address on the cache device is enabled by a first feature of the operating system, and wherein the operating system is configured to activate the first feature when it is determined that the operating system is being shut down.

18. The computer of claim 14, wherein the at least one processor is programmed to prevent an occurrence of a write operation to an address on the storage medium which is not also performed to an address on the cache device.

19. The computer of claim 18, wherein the write operation which the at least one processor is programmed to prevent is to a volume on the storage medium, and wherein the at least one processor is programmed to prevent the occurrence of the write operation to the address on the storage medium by attempting to make the volume inaccessible.

20. The computer of claim 14, the write operation which the at least one processor is programmed to identify is to data stored in a file system on the storage medium, and wherein the at least one processor is programmed to identify an occurrence of a write operation to an address on the storage medium by placing a log used by the file system in a particular state prior to reboot of the operating system, and determining whether the log is in the particular state following reboot of the operating system.