SCALABLE SECONDARY STORAGE SYSTEMS AND METHODS

• US 20100064166A1
• Filed: 07/29/2009
• Published: 03/11/2010
• Est. Priority Date: 09/11/2008
• Status: Active Grant

• Current Assignee: NEC Corporation
• Original Assignee: NEC Laboratories America Inc (NEC Corporation)
• Inventors: Dubnicki, Cezary, Ungureanu, Cristian
• Granted Patent: US 7,992,037 B2
• Time in Patent Office: Days
• US Class Current: 714/4
• CPC Class Codes: G06F 11/1076 Parity data used in redundant arrays of independent storages, e.g. in RAID systems
• CPC Class Codes: G06F 11/1453 using de-duplication of the data
• CPC Class Codes: G06F 16/10 File systems; File servers

Abstract

Exemplary systems and methods in accordance with embodiments of the present invention may provide a plurality of data services by employing splittable, mergable and transferable redundant chains of data containers. The chains and containers may be automatically split and/or merged in response to changes in storage node network configurations and may be stored in erasure coded fragments distributed across different storage nodes. Data services provided in a distributed secondary storage system utilizing redundant chains of containers may include global deduplication, dynamic scalability, support for multiple redundancy classes, data location, fast reading and writing of data and rebuilding of data due to node or disk failures.

Claims

1. A method for managing data on a secondary storage system comprising:

distributing data blocks to different data containers located in a plurality of different physical storage nodes in a node network to generate redundant chains of data containers in the nodes;

detecting an addition of active storage nodes to the network;

automatically splitting at least one chain of containers in response to detecting the addition; and

transferring at least a portion of data split from the at least one chain of containers from one of said storage nodes to another of said storage nodes to enhance system robustness against node failures.

2. The method of claim 1, wherein automatically splitting comprises splitting at least one data container and wherein the at least a portion of data is stored in the at least one data container prior to the splitting.

3. The method of claim 1, wherein automatically splitting comprises separating at least one data container from the at least one chain of containers and wherein said at least a portion of data is stored in the at least one data container prior to the splitting.

4. The method of claim 3, further comprising:

merging said at least one data container with another data container.

5. The method of claim 1, wherein at least one of said chains of data containers include the same metadata as other chains of data containers, wherein the metadata describes data block information.

6. The method of claim 5, further comprising:

rebuilding data in response to a failure using said metadata.

7. The method of claim 5, wherein the metadata include pointers between data blocks in the data containers.

8. The method of claim 7 further comprising:

deleting data using said pointers.

9. The method of claim 1, further comprising:

erasure coding said data blocks to generate erasure coded fragments, wherein said distributing comprises storing the erasure coded fragments in said different data containers such that fragments originating from one of said data blocks are stored on different storage nodes.

10. The method of claim 9, further comprising:

determining whether any of said redundant chains of data containers include holes to determine whether at least one of said data blocks is available in the secondary storage system.

11. A secondary storage system comprising:

a network of physical storage nodes, wherein each storage node includes

a storage medium configured to store fragments of data blocks in a chain of data containers that is redundant with respect to chains of data containers in other storage nodes; and

a storage server configured to detect an addition of active storage nodes to the network, to automatically split at least one chain of containers on said storage medium in response to detecting the addition, and to transfer at least a portion of data split from the at least one chain of containers to a different storage node to enhance system robustness against node failures.

12. The system of claim 11, wherein the storage server is further configured to perform at least one of:

data reading, data writing, determination of data availability, data transfer, distributed global duplicate elimination, data rebuilding, and data deletion.

13. The system of claim 11, wherein the at least one of said chains of data containers include the same metadata as other chains of data containers, wherein the metadata describes data block information.

14. The system of claim 13, wherein the metadata include pointers between data blocks in the data containers and wherein the storage server is configured to perform data deletion using said pointers.

15. The system of claim 11, wherein the fragments of data blocks are content-addressed on said storage medium in accordance with a hash function.

16. The system of claim 15, wherein different prefixes of content addresses are associated with a different subset of storage nodes.

17. The system of 16, wherein the automatic split comprises extending at least one of said prefixes to generate at least one additional subset of storage nodes.

18. The system of claim 17, wherein the transfer comprises distributing the at least a portion of data split from the at least one chain of containers to the additional subset.

19. A method for managing data on a secondary storage system comprising:

distributing data blocks to different data containers located in a plurality of different physical storage nodes in a node network to generate redundant chains of data containers in the nodes;

detecting a change in the number of active storage nodes in the network; and

automatically merging at least one data container located in one of said storage nodes with another data container located in a different storage node in response to detecting the change to ensure manageability of the containers.

20. The method of claim 19, wherein said change in the number of active nodes comprises at least one of:

an addition of at least one storage node to the network or a removal of at least one storage node from the network.