High performance journaling for replication and continuous data protection

• US 8,065,442 B1
• Filed: 11/19/2008
• Issued: 11/22/2011
• Est. Priority Date: 11/19/2007
• Status: Active Grant

• Current Assignee: AmZetta Technologies, LLC (American Megatrends International, LLC)
• Original Assignee: American Megatrends Incorporated
• Inventors: Maliakal, Jomy, Chatterjee, Paresh, Grandhi, Suresh, Subramanian, Srikumar
• Primary Examiner(s): Peyton; Tammara
• Application Number:
  US12/273,996
• Time in Patent Office: 1,098 Days
• Field of Search: 710/5, 710/52, 707/8, 707/201
• US Class Current: 710/5
• CPC Class Codes: G06F 11/1471 involving logging of persistent data for recovery
• CPC Class Codes: G06F 11/2066 Optimisation of the communication load
• CPC Class Codes: G06F 2201/81 Threshold

Abstract

Technologies are described herein for high-performance and space efficient journaling mechanisms. A journal queue can buffer incoming I/O operations. The queue can be read by a journal write module. The journal write module can form metadata headers to store into the journal along with each I/O operation. Compound metadata headers may be formed for multiple I/O operations and the combination of multiple I/O operations may be efficiently stored to the journal as one journal write. Multiple journal entries may be written into the journal in parallel while maintaining the sequential ordering of the journal. A queue depth threshold can be established for determining if journal entries should be combined into compound entries or not. Multiple concurrent readers can support continuous data protection, and various data replication features. Each reader can be represented as a reader pointer and a journal writer can invalidate reader pointers when related data is overwritten.

Claims

1. A method for journaling within a data storage system, the method comprising:

receiving an I/O operation;

queuing the received I/O operation in a journal queue;

supporting combining multiple I/O operations into a compound journal entry;

forming a compound metadata entry associated with each of the multiple I/O operation within the compound journal entry;

supporting a concurrent parallel write of multiple journal entries; and

establishing a threshold journal queue depth, and forming compound journal entries in response to a depth of the journal queue exceeding the threshold journal queue depth.

2. The method of claim 1, further comprising supporting multiple concurrent journal readers.

3. The method of claim 1, further comprising invalidating a reader pointer in response to overwriting the associated journal data.

4. The method of claim 3, wherein invalidating a reader pointer comprises applying a lightweight lock to the reader pointer.

5. The method of claim 3, further comprising transitioning a reader from journal assisted reads to snapshot assisted reads in response to the reader pointer being invalidated.

6. The method of claim 1, further comprising establishing a threshold journal queue depth, and forming single journal entries in response to a depth of the journal queue being below the threshold journal queue depth.

7. A computer storage medium having computer-executable instructions stored thereon which, when executed by a computer system, cause the computer system to:

receive an I/O operation;

queue the received I/O operation in a journal queue;

support combining multiple I/O operations into a compound journal entry;

form a compound metadata entry associated with each of the multiple I/O operation within the compound journal entry;

support a concurrent parallel write of multiple journal entries; and

establish a threshold journal queue depth, and forming compound journal entries in response to a depth of the journal queue exceeding the threshold journal queue depth.

8. The computer storage medium of claim 7, further comprising causing the computer system to support multiple concurrent journal readers.

9. The computer storage medium of claim 7, further comprising causing the computer system to invalidate a reader pointer in response to overwriting the associated journal data.

10. The computer storage medium of claim 9, wherein invalidating a reader pointer comprises applying a lightweight lock to the reader pointer.

11. The computer storage medium of claim 9, further comprising causing the computer system to transition a reader from journal assisted reads to snapshot assisted reads in response to the reader pointer being invalidated.

12. The computer storage medium of claim 7, further comprising causing the computer system to establish a threshold journal queue depth, and forming single journal entries in response to a depth of the journal queue being below the threshold journal queue depth.

13. A data storage system comprising:

a storage server;

a physical storage device associated with the storage server;

a processing unit associated with the storage server;

and one or more modules for execution on the processing unit, operable to receive an I/O operation, queue the received I/O operation in a journal queue, support combining multiple I/O operations into a compound journal entry, form a compound metadata entry associated with each of the multiple I/O operation within the compound journal entry, and support a concurrent parallel write of multiple journal entries,

wherein the one or more modules for execution on the processing unit are further operable to establish a threshold journal queue depth, and forming compound journal entries in response to a depth of the journal queue exceeding the threshold journal queue depth.

14. The data storage system of claim 13, wherein the one or more modules for execution on the processing unit are further operable to support multiple concurrent journal readers.

15. The data storage system of claim 13, wherein the one or more modules for execution on the processing unit are further operable to invalidate a reader pointer in response to overwriting the associated journal data.

16. The data storage system of claim 15, wherein invalidating a reader pointer comprises applying a lightweight lock to the reader pointer.

17. The data storage system of claim 15, wherein the one or more modules for execution on the processing unit are further operable to transition a reader from journal assisted reads to snapshot assisted reads in response to the reader pointer being invalidated.