MANAGING DATA ACCESS REQUESTS AFTER PERSISTENT SNAPSHOTS

• US 20120054152A1
• Filed: 08/26/2010
• Published: 03/01/2012
• Est. Priority Date: 08/26/2010
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Adkins, Janet E., Chang, Joon, Gjertsen, Robert K., Palsule, Ninad S.
• Granted Patent: US 8,306,950 B2
• Time in Patent Office: Days
• US Class Current: 707/623
• CPC Class Codes: G06F 16/2365 Ensuring data consistency and integrity

Abstract

A client node on which a file system is mounted can be configured to efficiently handle content access requests after a snapshot is created. The client node can maintain generation numbers at a data block, a file, and a fileset level to determine whether a data block is part of a snapshot generation and consequently whether the data block is writable in a current generation. Data block mappings and write access permissions associated with the data blocks need not be revoked at the client node prior to creating the snapshot. Cached data block mappings can be accessed to identify data blocks for servicing a read request. The data block can be updated in place if the data block is not part of the snapshot generation. The write request can be serviced at a new data block if the data block is part of the snapshot generation.

Claims

1. A method comprising:

associating, at a client node on which a clustered file system is mounted, a first generation number with one or more files of the clustered file system, wherein the first generation number indicates a state of the one or more files following a snapshot of the one or more files;

in response to a write request directed to a first data block associated with the one or more files,

determining whether the first data block is part of the snapshot of the one or more files based, at least in part, on the first generation number;

if the first data block is determined not to be part of the snapshot of the one or more files, updating the first data block in accordance with the write request; and

if the first data block is determined to be part of the snapshot of the one or more files,

transmitting, to a server that co-ordinates a plurality of client nodes on which the clustered file system is mounted, an allocation request for a second data block for servicing the write request, wherein the plurality of client nodes comprise the client node; and

updating the second data block in accordance with the write request.

2. The method of claim 1, wherein said determining whether the first data block is part of the snapshot of the one or more files in response to the write request directed to the first data block associated with the one or more files comprises:

determining a block generation number associated with the first data block;

determining that the first data block is associated with a first file of the one or more files;

determining a current file generation number associated with the first file;

determining a greater of the current file generation number associated with the first file and the first generation number;

comparing the block generation number against the greater of the current file generation number associated with the file and the first generation number;

determining that the first data block is not part of the snapshot if the block generation number is greater than the greater of the current file generation number and the first generation number; and

determining that the first data block is part of the snapshot if the block generation number is less than or equal to the greater of the current file generation number and the first generation number.

3. The method of claim 1, further comprising:

receiving the first generation number at the client node from the server;

determining that the client node comprises data not written to a storage disk associated with the clustered file system, in response to said receiving the first generation number;

flushing the data not written to the storage disk to synchronize the client node with the storage disk associated with the clustered file system; and

transmitting, to the server, metadata associated with the data flushed to the storage disk.

4. The method of claim 3, wherein said receiving the first generation number at the client node further comprises:

preventing purging of a cache associated with the client node to maintain data block mappings that indicate a mapping between logical data blocks of a file and physical data blocks on a storage disk and to maintain access permissions associated with the physical data blocks.

5. The method of claim 1, wherein said updating the second data block in accordance with the write request if the first data block is determined to be part of the snapshot of the one or more files comprises:

copying content of the first data block to the second data block;

updating the content of the second data block in accordance with the write request;

6. The method of claim 5, further comprising:

determining a new block generation number associated with the second data block;

updating metadata associated with the second data block to reflect the new block generation number associated with the second data block, wherein the new block generation number associated with the second data block indicates that the second data block is not part of the snapshot of the one or more files.

7. The method of claim 1, wherein said transmitting the allocation request for the second data block for servicing the write request if the first data block is determined to be part of the snapshot of the one or more files, comprises:

identifying a data block mapping that maps a logical data block to the first data block; and

updating the data block mapping to map the logical data block to the second data block.

8. The method of claim 1, further comprising:

in response to a read request directed to a third data block associated with the one or more files, accessing the third data block based on data block mappings maintained by the client node.

9. The method of claim 8, wherein said accessing the third data block based on the data block mappings maintained by the client node comprises:

receiving the read request to access content of a file;

determining a logical data block that constitutes the file; and

determining, from the data block mappings maintained by the client node, the third data block of a storage disk associated with the clustered file system that maps to the logical data block that constitutes the file.

10. The method of claim 1, further comprising:

updating a predetermined memory location to replace a previous generation number associated with the one or more files and to store the first generation number.

11. A computer program product for managing data access requests after a persistent snapshot, the computer program product comprising:

a computer readable storage medium having computer usable program code embodied therewith, the computer readable program code configured to;

associate a first generation number with one or more files of a clustered file system, wherein the first generation number indicates a state of the one or more files following creation of a snapshot of the one or more files;

in response to a write request directed to a first data block associated with the one or more files,

determine whether the first data block is part of the snapshot of the one or more files based, at least in part, on the first generation number;

if the first data block is determined not to be part of the snapshot of the one or more files, update the first data block in accordance with the write request; and

if the first data block is determined to be part of the snapshot of the one or more files,

transmit, to a server that co-ordinates a plurality of client nodes on which the clustered file system is mounted, an allocation request for a second data block for servicing the write request; and

update the second data block in accordance with the write request.

12. The computer program product of claim 11, wherein the computer readable program code configured to determine whether the first data block is part of the snapshot of the one or more files in response to the write request directed to the first data block associated with the one or more files comprises the computer readable program code configured to:

determine a block generation number associated with the first data block;

determine that the first data block is associated with a first file of the one or more files;

determine a current file generation number associated with the first file;

determine a greater of the current file generation number associated with the first file and the first generation number; and

compare the block generation number against the greater of the current file generation number associated with the first file and the first generation number;

determine that the first data block is not part of the snapshot if the block generation number is greater than the greater of the current file generation number and the first generation number; and

determine that the first data block is part of the snapshot if the block generation number is less than or equal to the greater of the current file generation number and the first generation number.

13. The computer program product of claim 11, wherein the computer readable program code is further configured to:

receive the first generation number at the client node from the server;

determine that the client node comprises data not written to a storage disk associated with the clustered file system, in response to the computer readable program code receiving the first generation number;

flush the data not written to the storage disk to synchronize the client node with the storage disk associated with the clustered file system; and

transmit, to the server, metadata associated with the data flushed to the storage disk.

14. The computer program product of claim 11, wherein the computer readable program code is further configured to:

determine a new block generation number associated with the second data block;

updating metadata associated with the second data block to reflect the new block generation number associated with the second data block, wherein the new block generation number associated with the second data block indicates that the second data block is not part of the snapshot of the one or more files.

15. The computer program product of claim 11, wherein the computer readable program code configured to update the second data block in accordance with the write request if the first data block is determined to be part of the snapshot of the one or more files comprises the computer readable program code configured to:

copy content of the first data block to the second data block;

update the content of the second data block in accordance with the write request;

16. The computer program product of claim 11, wherein the computer readable program code configured to transmit the allocation request for the second data block for servicing the write request if the first data block is determined to be part of the snapshot of the one or more files, comprises the computer readable program code configured to:

identify a data block mapping that maps a logical data block to the first data block; and

update the data block mapping to map the logical data block to the first data block.

17. The computer program product of claim 11, wherein the computer readable program code is further configured to:

in response to a read request directed to a third data block associated with the one or more files, access the third data block based on data block mappings maintained by the client node.

18. A machine comprising:

a processor;

a network interface coupled with the processor, the network interface operable to receive, at the machine on which a clustered file system is mounted, a first generation number associated with one or more files of the clustered file system, wherein the first generation number indicates a state of the one or more files following creation of a snapshot of the one or more files; and

a generation number comparison unit operable to;

indicate the first generation number for the one or more files;

in response to a write request directed to a first data block associated with the one or more files, determine whether the first data block is part of the snapshot of the one or more files;

indicate to a read/write unit whether or not the first data block is part of the snapshot;

the read/write unit operable to;

if the generation number comparison unit indicates that the first data block is not part of the snapshot, update the first data block in accordance with the write request; and

if the generation number comparison unit indicates that the first data block is part of the snapshot,

transmit, to a server that co-ordinates a plurality of machines on which the clustered file system is mounted, an allocation request for a second data block for servicing the write request, wherein the plurality of machines comprise the machine; and

update the second data block in accordance with the write request.

19. The machine of claim 18, wherein the generation number comparison unit operable to determine whether the first data block is part of the snapshot of the one or more files in response to the write request directed to the first data block associated with the one or more files comprises the generation number comparison unit operable to:

determine a block generation number associated with the first data block;

determine that the first data block is associated with a first file of the one or more files;

determine a current file generation number associated with the first file;

determine a greater of the current file generation number associated with the first file and the first generation number;

compare the block generation number against the greater of the current file generation number associated with the first file and the first generation number;

determine that the first data block is not part of the snapshot if the block generation number is greater than the greater of the current file generation number and the first generation number; and

determine that the first data block is part of the snapshot if the block generation number is less than or equal to the greater of the current file generation number and the first generation number.

20. The machine of claim 18, wherein the generation number comparison unit and the read/write unit comprise the computer readable storage medium.