Content addressable storage system

• US 8,560,503 B1
• Filed: 01/26/2006
• Issued: 10/15/2013
• Est. Priority Date: 01/26/2006
• Status: Active Grant

• Current Assignee: NetApp, Inc.
• Original Assignee: NetApp, Inc.
• Inventors: McManis, Charles
• Primary Examiner(s): CHANNAVAJJALA, SRIRAMA T
• Application Number:
  US11/339,888
• Time in Patent Office: 2,819 Days
• Field of Search: 707/640, 707/664, 707/692, 707693-694, 707/796, 707/812, 711113-114, 711136-137, 711142-143, 711159-160, 711/205
• US Class Current: 707/664
• CPC Class Codes: G06F 13/00 Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units interface circuits for specific input/output devices G06F3/00 ; multiprogram control therefor G06F9/46; multiprocessor systems G06F15/16
• CPC Class Codes: G06F 13/4234 being a memory bus
• CPC Class Codes: G06F 16/00 Information retrieval; Database structures therefor; File system structures therefor
• CPC Class Codes: G06F 17/00 Digital computing or data processing equipment or methods, specially adapted for specific functions information retrieval, database structures or file system structures therefor G06F16/00

Abstract

A content addressable storage system (CASS) comprises a storage system coupled to one or more content addressable storage array elements (CASAEs). The storage system is coupled to storage resources of a local storage array configured to store a high level meta-directory that describes data containers, e.g., directories, files and logical units (luns), served by the CASS. In addition, each CASAE is coupled to storage resources of a remote storage array configured to store user data (data blocks) of the data containers served by the CASS. The CASAE is configured to perform a content addressable storage computation on each data block written to its storage resources in order to prevent storage of two or more blocks with the same data. Notably, the CASAE interacts with its associated remote storage array to form a CASAE “storage brick”.

Claims

1. A content addressable storage system, comprising:

a storage system configured to connect to a local storage array, the local storage array configured to store a meta-directory that describes a data container served by the content addressable storage system;

a first content addressable storage array element of a plurality of content addressable storage array elements configured to connect to the storage system and a remote storage array;

the remote storage array configured to store data blocks of the data container served by the content addressable storage system and described by the meta-directory;

the first content addressable storage array element further configured to interact with the remote storage array to form a storage brick; and

the first content addressable storage array element further configured determine, before the data blocks are stored, whether the data blocks to be stored match blocks of data previously stored at the remote storage array.

2. The content addressable storage system of claim 1 wherein the data container comprises one of a file, a directory, and a logical unit.

3. The content addressable storage system of claim 1 wherein a plurality of storage bricks is organized as a group and operated as a RAID group.

4. The system of claim 1 further comprising:

a switching fabric configured to couple the storage system to the first content addressable storage array element.

5. A method for providing a content addressable storage system, comprising:

coupling a storage system to a first content addressable storage array element (CASAE) of a plurality of content addressable storage array elements, and coupling the storage system to a network port of the first CASAE;

storing a meta-directory on a local storage array of the storage system, the meta-directory describing a data container served by the content addressable storage system;

storing data blocks of the data container on a remote storage array operatively connected to the first CASAE, the data blocks of the data container served by the content addressable storage system and described by the meta-directory; and

performing, on at least one of the plurality of CASAEs, a determination before the data blocks are stored of whether the data blocks to be stored match previously stored blocks of data at one or more remote storage arrays.

6. The method of claim 5 further comprising forming a storage brick through an interaction between the first CASAE and the remote storage array.

7. The method of claim 6 further comprising coupling the storage system to a second storage brick to allow data striping across the storage bricks.

8. The method of claim 7 further comprising operating the storage bricks in parallel to allow aggregation of resources among the storage bricks, thereby increasing performance of content addressable storage computations.

9. The method of claim 8 further comprising:

organizing the storage bricks as a group; and

operating the group as a RAID group.

10. The method of claim 9 further comprising designating one of the storage bricks as a parity brick to thereby enable storage resources of the parity brick to function as parity storage resources in a predetermined RAID implementation.

11. The method of claim 7 further comprising:

receiving a write request at the storage system;

passing the write request to a file system executing on the storage system;

constructing a block layout message that identifies blocks to be used for storing user write data associated with the write request;

inserting a first block number into the block layout message;

passing the block layout message to a RAID system of the storage system, and attempting, by the RAID system, to balance the write data across the storage bricks; and

forwarding the block layout message, a write operation, and the write data to at least one of the first and second storage bricks.

12. The method of claim 11 further comprises determining a location at which to store the write data in a logical unit of the at least one storage brick.

13. The method of claim 12 further comprising:

returning a physical block number to the file system, the physical block number indicating the location at which the write data resides in a data block of the logical unit;

performing an operation to change the first block number to the physical block number; and

updating at least one buffer tree pointer for the logical unit in the local storage array to reference the physical block number.

14. Apparatus configured to provide a content addressable storage system, comprising:

means for coupling a storage system to a first content addressable storage array element (CASAE) of a plurality of content addressable storage array elements, and coupling the storage system to a network port of the first CASAE;

means for storing a meta-directory on a local storage array of the storage system, the meta-directory describing a data container served by the content addressable storage system;

means for storing data blocks of the data container on a remote storage array operatively connected to the first CASAE, the data blocks of the data container served by the content addressable storage system and described by the meta-directory; and

means for performing, on at least one of the plurality of CASAEs, a determination before the data blocks are stored of whether the data blocks to be stored match previously stored blocks of data at one or more remote storage arrays.

15. The apparatus of claim 14 further comprising means for forming a storage brick comprising the first CASAE and the remote storage array.

16. The apparatus of claim 15 further comprising means for coupling the storage system to a second storage brick to allow data striping across the storage bricks.

17. The apparatus of claim 16 further comprising means for operating the storage bricks in parallel to allow aggregation of resources among the storage bricks, thereby enabling increased performance of content addressable storage computations.

18. A computer readable medium storing executable program instructions for execution by a processor, comprising:

program instructions that couple a storage system to a first content addressable storage array element (CASAE) of a plurality of content addressable storage array elements, and coupling the storage system to a network port of the first CASAE;

program instructions that store a meta-directory on a local storage array of the storage system, the meta-directory describing a data container served by the content addressable storage system;

program instructions that store data blocks of the data container on a remote storage array operatively connected to the first CASAE, the data blocks of the data container served by the content addressable storage system and described by the meta-directory; and

program instructions that perform, on at least one of the plurality of CASAEs, a determination before the data blocks are stored of whether the data blocks to be stored match previously stored blocks of data at one or more remote storage arrays.

19. The computer readable medium of claim 18 wherein the program instructions that perform comprise program instructions that determine a location at which to store write data in a logical unit of the first CASAE.

20. The computer readable medium of claim 19 further comprising program instructions that:

return a physical block number to the storage system, the physical block number indicating the location at which the write data resides in a data block of the logical unit;

perform a fix-up operation to change a special block number to the physical block number; and

update at least one buffer tree pointer for the logical unit in the local storage array to reference the physical block number.

21. A method for operating a storage system, comprising:

coupling the storage system to a plurality of content addressable storage array elements (CASAEs) over a network, the storage system coupled to a switching fabric which is coupled to a network port on each CASAE;

storing a meta-directory of inodes pointing to data blocks on a locally coupled storage array of the storage system; and

performing, by a processor on each CASAE at the lun level, a computation on a plurality of data blocks stored by each CASAE to speed the computation and to provide parallel access of the data blocks and the meta-directory, wherein each CASAE of the plurality of CASAEs performs the computation in parallel with each other CASAE.

22. The method of claim 21 further comprising:

forming a storage brick through an interaction between a first CASAE of the plurality of CASAEs and a remote storage array.

23. The method of claim 22 further comprising:

forming a plurality of storage bricks; and

striping data across the storage bricks.

24. The method of claim 23 further comprising:

organizing the storage bricks as a group; and

operating the group as a RAID group.

25. The method of claim 23 further comprising:

designating one of the storage bricks as a parity brick to thereby enable storage of parity.

26. The method of claim 23 further comprising:

receiving a write request at the storage system;

passing the write request to a file system executing on the storage system;

constructing a block layout message that identifies data blocks to for storing user write data associated with the write request;

inserting a special block number into the block layout message;

passing the block layout message to a RAID system of the storage system, and attempting by the RAID system to balance the write data across the storage bricks; and

forwarding the block layout message, a write operation and the write data, to at least one of the storage bricks.

27. The method of claim 26 further comprising:

determining a location at which to store the write data in a logical unit of the at least one storage brick.

28. The method of claim 27 further comprising:

returning a physical block number to the file system, the physical block number indicating the location at which the write data resides in a data block of the logical unit;

performing a fix-up operation to change the special block number to the physical block number; and

updating at least one buffer tree pointer for the logical unit in the local storage array to reference the physical block number of the data block.

29. A non-transitory computer readable medium storing executable program instructions executed by a processor, comprising:

program instructions that couple the storage system to a plurality of content addressable storage array elements (CASAEs) over a network, the storage system coupled to a switching fabric which is coupled to a network port on each CASAE;

program instructions that store a meta-directory of inodes pointing to data blocks on a locally coupled storage array of the storage system; and

program instructions that perform, by a processor on each CASAE at the lun level, a computation on a plurality of data blocks stored by each CASAE to speed the computation and to provide parallel access of the data blocks and the meta-directory, wherein each CASAE of the plurality of CASAEs performs the computation in parallel with each other CASAE.