System and method to preserve and recover unwritten data present in data cache of a disk subsystem across power outages

• US 8,347,041 B2
• Filed: 01/02/2009
• Issued: 01/01/2013
• Est. Priority Date: 01/02/2009
• Status: Expired due to Fees

• Current Assignee: Avago Technologies International Sales Pte Limited (Broadcom, Inc.)
• Original Assignee: LSI Corporation (Broadcom, Inc.)
• Inventors: Banerjee, Arindam
• Primary Examiner(s): Dudek, Jr., Edward
• Assistant Examiner(s): Fishburn, John P
• Application Number:
  US12/348,142
• Publication Number:
  US 20100174870A1
• Time in Patent Office: 1,460 Days
• Field of Search: None
• US Class Current: 711/143
• CPC Class Codes: G06F 1/30 Means for acting in the event of power-supply failure or interruption, e.g. power-supply fluctuations for resetting only G06F1/24
• CPC Class Codes: G06F 12/0866 for peripheral storage systems, e.g. disk cache
• CPC Class Codes: G06F 2212/1028 Power efficiency
• CPC Class Codes: G06F 2212/214 Solid state disk
• CPC Class Codes: Y02D 10/00 Energy efficient computing, e.g. low power processors, power management or thermal management

Abstract

Disclosed are a system and method to preserve and recover unwritten data present in data cache of a disk subsystem across power outages. In one embodiment, a method of a controller is described. The method includes applying a write-back technique between a host server and a data store, accessing a dirty data in a cache memory during a power outage. The method may apply an algorithm for efficiently offloading the dirty data to a non-volatile storage device during the power outage. In addition the method may apply the algorithm to efficiently transfer the dirty data from the non-volatile storage device to the data store when power is restored.

Claims

1. A method of a controller coupled between a host server and a data store comprising:

solely packing blocks of a cache memory of the controller corresponding to a dirty data in an order of a volume number and a logical block address within a volume, the dirty data being data from the host server prior to being written to the data store;

backing up the packed blocks of the cache memory corresponding to the dirty data in a intermediary backup non-volatile storage device distinct from both the cache memory and the data store and storing packing information in the intermediary backup non-volatile storage device, with storing of the packing information at the cache memory being avoided;

maintaining, in the packing information, a state of the packed blocks of the cache memory corresponding to the dirty data immediately prior to a power outage during the power outage; and

sequentially accessing the packed blocks of the cache memory corresponding to the dirty data in the intermediary backup non-volatile storage device based on the packing information to transfer the dirty data to the data store following restoration of power.

2. The method of claim 1, wherein when there is a plurality of intermediary backup non-volatile storage devices, the method further comprises:

ensuring parallel access to each of the plurality of intermediary backup non-volatile storage devices during backup of the dirty data and the transfer thereof to the data store.

3. The method of claim 2, further comprising creating one of:

a volume group and a disk group based on the plurality of intermediary backup non-volatile storage devices.

4. The method of claim 2, further comprising utilizing a data striping technique to achieve parallelism.

5. The method of claim 3, further comprising creating logical volumes having a portion of backup data on each of the plurality of intermediary backup non-volatile storage devices constituting the volume group or the disk group during a backup operation.

6. The method of claim 1, wherein solely packing the blocks of the cache memory corresponding to the dirty data in the order of the volume number and the logical block address comprises packing the blocks from different physical locations in the cache memory in the order.

7. The method of claim 1, further comprising:

recreating the packing information during every period of interruption during at least one of a backup and a restore operation; and

locating a physical location of a block of the cache memory corresponding to the dirty data within the intermediary backup non-volatile storage device using the packing information.

8. The method of claim 1, further comprising:

transferring the dirty data from the intermediary backup non-volatile storage device to the data store based on a cache synchronization operation.

9. The method of claim 1, further comprising:

ensuring that the packed blocks of the cache memory corresponding to the dirty data and associated with interrupted write operations are laid out contiguously within the intermediary backup non-volatile storage device.

10. The method of claim 1, further comprising:

allowing the transfer of the dirty data to the data store in a context of an Input/Output (IO) initiated by the host server.

11. The method of claim 1, further comprising:

invalidating the dirty data within the intermediary backup non-volatile storage device after the transfer of the dirty data to the data store.

12. The method of claim 11, wherein the invalidation comprises:

updating specific fields within the packing information stored within the intermediary backup non-volatile storage device; and

avoiding an expensive read-modify-write operation during the invalidation by utilizing a bitmap technique.

13. A system comprising:

a host server;

a data store;

a controller comprising a cache memory configured to include dirty data corresponding to data from the host server prior to being written to the data store; and

a intermediary backup non-volatile storage device distinct from both the data store and the cache memory,

wherein the controller is configured to enable;

solely packing blocks of the cache memory corresponding to the dirty data in an order of a volume number and a logical block address within a volume;

backing up the packed blocks of the cache memory corresponding to the dirty data in the intermediary backup non-volatile storage device and storing packing information therein, with storing of the packing information at the cache memory being avoided;

maintaining, in the packing information, a state of the packed blocks of the cache memory corresponding to the dirty data immediately prior to a power outage during the power outage; and

sequentially accessing the packed blocks of the cache memory corresponding to the dirty data in the intermediary backup non-volatile storage device based on the packing information to transfer the dirty data to the data store following restoration of power.

14. The system of claim 13, wherein when the system comprises a plurality of intermediary backup non-volatile storage devices, the controller is further configured to ensure parallel access to each of the plurality of intermediary backup non-volatile storage devices during backup of the dirty data and the transfer thereof to the data store.

15. The system of claim 14, wherein the controller is further configured to enable creation of one of:

a volume group and a disk group based on the plurality of intermediary backup non-volatile storage devices.

16. The system of claim 13, wherein the controller is further configured to enable:

recreating the packing information during every period of interruption during at least one of a backup and a restore operation; and

locating a physical location of a block of the cache memory corresponding to the dirty data within the intermediary backup non-volatile storage device using the packing information.

17. The system of claim 13, wherein the controller is further configured to enable:

transferring the dirty data from the intermediary backup non-volatile storage device to the data store based on a cache synchronization operation.

18. The system of claim 13, wherein the controller is further configured to enable:

allowing the transfer of the dirty data to the data store in a context of an Input/Output (IO) initiated by the host server.

19. A method of a controller coupled between a host server and a data store comprising:

solely packing blocks of a cache memory of the controller corresponding to a dirty data in an order of a volume number and a logical block address within a volume, the dirty data being data from the host server prior to being written to the data store;

backing up the packed blocks of the cache memory corresponding to the dirty data in a plurality of intermediary backup non-volatile storage devices distinct from both the cache memory and the data store and storing packing information in the plurality of intermediary backup non-volatile storage devices, with storing of the packing information at the cache memory being avoided;

maintaining, in the packing information, a state of the packed blocks of the cache memory corresponding to the dirty data immediately prior to a power outage during the power outage;

sequentially accessing the packed blocks of the cache memory corresponding to the dirty data in the plurality of intermediary backup non-volatile storage devices based on the packing information to transfer the dirty data to the data store following restoration of power; and

ensuring parallel access to each of the plurality of intermediary backup non-volatile storage devices during backup of the dirty data and the transfer thereof to the data store.

20. The method of claim 19, further comprising creating one of:

a volume group and a disk group based on the plurality of intermediary backup non-volatile storage devices.