Ditto blocks

• US 7,743,225 B2
• Filed: 04/19/2006
• Issued: 06/22/2010
• Est. Priority Date: 11/04/2005
• Status: Active Grant

• Current Assignee: Oracle America, Inc. (Oracle Corporation)
• Original Assignee: Oracle America, Inc. (Oracle Corporation)
• Inventors: Ahrens, Matthew A., Bonwick, Jeffrey S., Moore, William H.
• Primary Examiner(s): Tsai; Sheng-Jen
• Assistant Examiner(s): Chrzanowski; Matthew R
• Application Number:
  US11/406,590
• Publication Number:
  US 20070106862A1
• Time in Patent Office: 1,525 Days
• Field of Search: None
• US Class Current: 711/162
• CPC Class Codes: G06F 11/1435 using file system or storage system metadata
• CPC Class Codes: G06F 11/2056 by mirroring
• CPC Class Codes: G06F 3/061 Improving I/O performance
• CPC Class Codes: G06F 3/0644 Management of space entities, e.g. partitions, extents, pools
• CPC Class Codes: G06F 3/065 Replication mechanisms
• CPC Class Codes: G06F 3/0683 Plurality of storage devices

Abstract

A method for storing a data block in a storage pool, including receiving a write operation for the data block, writing the data block into the storage pool, determining a first indirect block to store in the storage pool, wherein the first indirect block references the data block, writing the first indirect block to a first location in the storage pool, creating a copy of the first indirect block, and writing the copy of first indirect block to a second location in the storage pool.

Claims

1. A method for storing a data block in a storage pool, comprising:

receiving a write operation for the data block;

writing the data block into the storage pool;

determining a first indirect block to store in the storage pool, wherein the first indirect block comprises a first block pointer that references the data block;

writing the first indirect block to a first location in the storage pool;

creating a copy of the first indirect block;

writing the copy of first indirect block to a second location in the storage pool, wherein the copy of the first indirect block comprises a second block pointer that references the data block; and

writing a second indirect block into the storage pool, wherein the second indirect block comprises a third block pointer that references both the first indirect block and the copy of the first indirect block.

2. The method of claim 1, further comprising:

computing a number of physical blocks to obtain an allocation size for the first indirect block.

3. The method of claim 2, further comprising:

allocating rows corresponding to the allocation size to obtain a set of allocated rows for writing the first indirect block, wherein writing the first indirect block comprises writing the first indirect block to the set of allocated rows column-first.

4. The method of claim 1, wherein the data block, the first indirect block, and the copy of the first indirect block correspond to a portion of a hierarchical tree structure representing a file.

5. The method of claim 4, wherein the first indirect block and the copy of the first indirect block store metadata associated with the file.

6. A system, comprising:

a storage pool comprising;

a plurality of child blocks, wherein each of the plurality of child blocks comprises one selected from the group consisting of a data block and an indirect block, wherein each of the indirect blocks references at least one of the plurality of child blocks; and

a parent block referencing at least one indirect block; and

a storage pool allocator configured to store the parent block and the plurality of child blocks,

wherein the at least one indirect block is stored in a first location and a copy of the at least one indirect block is stored in a second location in the storage pool,

wherein the at least one indirect block references a data block,

wherein the data block is one of the plurality of child blocks,

wherein the copy of the at least one indirect block references the data block, and

wherein the first location and the second location are stored in a single block pointer of the parent block referencing the at least one indirect block.

7. The system of claim 6, wherein the storage pool comprises a plurality of disks for storing the plurality of child blocks and the parent block.

8. The system of claim 7, wherein the first location resides on a first one of the plurality of disks and the second location resides on a second one of the plurality of disks.

9. A method for reading data in a storage pool, comprising:

obtaining a first location of a first indirect block from a block pointer of a parent block;

determining whether the first indirect block can be retrieved from the first location;

when the first indirect block cannot be retrieved from the first location;

obtaining a second location of the first indirect block from the block pointer of the parent block, wherein the second location comprises a copy of the first indirect block; and

retrieving the first indirect block from the second location, and

retrieving the first indirect block from the first location, when the first indirect block can be retrieved from the first location,

wherein the first indirect block comprises a first block pointer referencing a child block, and the copy of the first indirect block comprises a second block pointer referencing the child block.

10. The method of claim 9, wherein the child block comprises one selected from the group consisting of a data block and a second indirect block.

11. The method of claim 10, further comprising:

when the first and second block pointer reference the data block;

retrieving the at least one data block using one of the first block pointer in the first indirect block and the second block pointer in the copy of the first indirect block; and

extracting and presenting data stored in the data block to a requesting process.

12. The method of claim 10, further comprising:

when the first and second block pointer reference the second indirect block and when the first indirect block cannot be retrieved from the first location;

obtaining a first location of the second indirect block from the second block pointer of the copy of the first indirect block;

determining whether the second indirect block can be retrieved from the first location;

when the second indirect block cannot be retrieved from the first location;

obtaining a second location of the second indirect block from the second block pointer of the copy of the first indirect block; and

retrieving the second indirect block from the second location, and

retrieving the second indirect block from the first location, when the second indirect block can be retrieved from the first location.

13. The method of claim 9, wherein the first location resides on a first disk and the second location resides on a second disk.

14. The method of claim 9, wherein the indirect block stores metadata associated with a file.

15. The method of claim 9, wherein the first indirect block and the copy of the first indirect block store metadata associated with the file.