Block-level Access to Parallel Storage

• US 20140068224A1
• Filed: 08/30/2012
• Published: 03/06/2014
• Est. Priority Date: 08/30/2012
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Nightingale, Edmund Bernard, Elson, Jeremy E., Mickens, James W., Fan, Bin, Kadav, Asim, Rashid, Richard F.
• Granted Patent: US 9,778,856 B2
• Time in Patent Office: Days
• US Class Current: 711/206
• CPC Class Codes: G06F 3/0613 in relation to throughput
• CPC Class Codes: G06F 3/064 Management of blocks
• CPC Class Codes: G06F 3/067 Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]

Abstract

The subject disclosure is directed towards one or more parallel storage components for parallelizing block-level input/output associated with remote file data. Based upon a mapping scheme, the file data is partitioned into a plurality of blocks in which each may be equal in size. A translator component of the parallel storage may determine a mapping between the plurality of blocks and a plurality of storage nodes such that at least a portion of the plurality of blocks is accessible in parallel. Such a mapping, for example, may place each block in a different storage node allowing the plurality of blocks to be retrieved simultaneously and in its entirety.

Claims

1. In a computing environment, a method performed at least in part on at least one processor, comprising, parallelizing input/output operations associated with file data, including, partitioning the file data into a plurality of blocks, identifying a plurality of storage nodes for the plurality of blocks, the plurality of storage nodes associated with parallel storage, translating addresses corresponding to the plurality of blocks into block-level addresses corresponding to the plurality of storage nodes, and using the block-level addresses to access one or more block sequences of the plurality of blocks in parallel when handling the input/output operations.

2. The method of claim 1, wherein translating addresses corresponding to the plurality of blocks further comprises mapping the block sequence to at least one corresponding location within one or more portions of the plurality of storage nodes.

3. The method of claim 2, wherein mapping the block sequence further comprises mapping the block sequence to random locations within the one or more portions of the plurality of storage nodes.

4. The method of claim 2 further comprising repeating the mapping step for each block sequence of the plurality of blocks.

5. The method of claim 2 further comprising modifying a size of at least a portion of the block sequence.

6. The method of claim 1 further comprising communicating the plurality of blocks to a computing device at a rate equal to at least a local data storage rate.

7. The method of claim 1 further comprising generating a file data representation comprising the block-level addresses of the one or more block sequences.

8. The method of claim 7 further comprising providing the file data representation to a computing device that uses the file data representation to initiate file system operations over a local area network.

9. The method of claim 1, wherein translating addresses corresponding to the plurality of blocks further comprises deterministically addressing the one or more block sequences.

10. The method of claim 1 further comprising modifying a mapping scheme based upon statistics associated with the handling the input/output operations.

11. The method of claim 1 further comprising communicating the plurality of blocks to recover the file data for a computing device.

12. In a computing environment, a system comprising, one or more parallel storage components coupled to a storage node group, including, a translator component configured to map virtual storage extent portions to an address space associated with the storage node group, wherein adjacent portions map to non-adjacent block-level addresses within the address space, and to retrieve at least a portion of the adjacent portions concurrently.

13. The system of claim 12, wherein the translator component is further configured to modify a mapping scheme for converting byte offsets within a virtual storage extent into block offsets within the address space.

14. The system of claim 12, wherein the translator component is further configured to determine a block size for a virtual storage extent and a number of tracts served by each storage node in the storage node group.

15. The system of claim 14, wherein the translator component is further configured to compute a tract identifier and a tract offset for each virtual storage extent offset, wherein the translator component combines the tract identifier and the tract offset to generate at least a portion of a block-level address.

16. The system of claim 12 further comprising a plurality of computing devices communicably coupled, within a physical boundary, to the storage node group, wherein the plurality of computing devices and the storage node group communicate data over a full bi-sectional bandwidth network.

17. The system of claim 12 wherein the translator component is further configured to (a) map a first set of blocks to a set of first locations amongst the plurality of storage nodes and to repeat step (a) for each next set of blocks.

18. One or more computer-readable media having computer-executable instructions stored thereon, which cause a computer to perform steps comprising:

accessing a file data representation comprising block-level addresses within a binary large object that corresponds to byte sequences comprising file data;

processing file system input/output operations that are directed to file data associated with the byte sequences, including transforming the file-level input/output operations into parallelized, block-level input/output operations; and

communicating the parallelized, block-level input/output operations to one or more storage nodes associated with the binary large object.

19. The one or more computer-readable media of claim 18 having further computer-executable instructions that cause the computer to perform steps comprising:

providing an interface for the one or more storage nodes, wherein a file system driver uses the interface to network boot a computing device or mount a file system over a virtual storage extent within the binary large object and initiates the file-level input/output operations on behalf of an unmodified software application running on the computing device.

20. The one or more computer-readable media of claim 19 having further computer-executable instructions that cause the computer to perform further steps comprising:

retrieving at least one contiguous byte sequence using block-level addresses associated with the virtual storage extent; and

restoring the virtual storage extent using the at least one contiguous byte sequence.