Storage system using fast storage and log-structured storage

• US 20040148462A1
• Filed: 01/27/2003
• Published: 07/29/2004
• Est. Priority Date: 01/27/2003
• Status: Active Grant

• Current Assignee: Hewlett Packard Enterprise Development LP (Hewlett-Packard Enterprise Company), Valtrus Innovations Limited (f/k/a Dolya Holdco 9 Limited) (Key Patent Innovations Limited)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Merchant, Arif, Uysal, Mustafa, Alvarez, Guillermo A.
• Granted Patent: US 7,073,029 B2
• Time in Patent Office: Days
• US Class Current: 711/118
• CPC Class Codes: G06F 11/1076 Parity data used in redundant arrays of independent storages, e.g. in RAID systems
• CPC Class Codes: G06F 2211/1038 LFS, i.e. Log Structured File System used in RAID systems with parity

Abstract

A computer storage system includes a controller and a storage device array. The storage device array includes a first sub-array and a fast storage device sub-array. The first sub-array includes one or more log-structured storage devices storing data. The fast storage device sub-array includes one or more fast storage devices storing a copy of the data stored in the first sub-array.

Claims

1. A computer storage system comprising:

a controller configured to perform data operations on a storage device array; and

the storage device array comprises a first sub-array and a fast storage sub-array storing data redundant to data stored in the first sub-array, wherein the first sub-array includes at least one log-structured storage device and the fast storage sub-array includes at least one fast storage device.

2. The computer storage system of claim 1, wherein the at least one log-structured storage device is divided into extents and the controller is operable to use a log-structured algorithm to write data into at least one of the extents.

3. The computer storage system of claim 1 further comprising a cache, wherein the log-structured algorithm comprises appending the data into the at least one extent and updating metadata in the cache associated with the data.

4. The computer storage system of claim 2, wherein the controller is operable to clean an extent of the at least one log-structured storage device, the extent to be cleaned including valid data.

5. The computer storage system of claim 4, wherein the controller is operable to read data corresponding to the valid data in an extent being cleaned from the fast storage sub array and append the data into at least one extent in the first sub-array using the log-structured algorithm.

6. The computer storage system of claim 4, wherein the controller is operable to mark the extent being cleaned as clean after all the valid data in that extent is appended to the at least one extent.

7. The computer storage system of claim 1, wherein the controller attempts to perform a read operation on the fast storage sub-array before attempting to perform the read operation on the first storage sub-array.

8. The computer storage system of claim 1, wherein the controller is configured to write data to the first sub-array in a log-structured layout and write the redundant data to the fast storage device sub-array.

9. The computer storage system of claim 8, wherein the redundant data comprises at least a copy of the data stored in the first sub-array.

10. The computer storage system of claim 1, wherein the fast storage sub-array comprises one or more of MEMS, flash RAM, magneto-resistive RAM.

11. The computer storage system of claim 1, wherein the first storage sub-array comprises one or more disks.

12. The computer storage system of claim 1, wherein new data written to the log-structured storage device is appended to the log-structured storage device.

13. In a computer storage system including at least one log-structured storage device and at least one fast storage device storing data redundant to data stored in the at least one log-structured storage device, wherein the at least one log-structured storage device is divided into extents for storing data, a method for cleaning an extent comprises steps of:

identifying valid data in an extent being cleaned in the log-structured storage device;

retrieving a copy of the valid data from the at least one fast storage device; and

appending the copy of the valid data to a current extent in the at least one log-structured storage device.

14. The method of claim 13, wherein the step of retrieving comprises steps of:

detecting an idle time for the at least one fast storage device; and

retrieving the copy of the valid data from the at least one fast storage device to a buffer during the idle time for the at least one fast storage device.

15. The method of claim 13, wherein the step of appending comprises steps of:

appending the copy of the valid data from the buffer to the current extent in the at least one log-structured storage device.

16. The method of claim 13, wherein the current extent comprises one or more of a current extent in the at least one log-structured storage device, a clean extent nearest to the previous current extent, and an extent with a greatest number of invalid blocks at the time the previous current extent became full.

17. The method of claim 13, wherein the computer storage system comprises a first storage sub-array including the at least one log-structured storage device and a fast storage sub-array including the at least one fast storage device.

18. A method of performing data operations in a storage system comprising at least one log-structured storage device and at least one fast storage device, the method comprising steps of:

receiving a request to perform a data operation;

determining whether the request provokes a write operation on the storage system; and

in response to the request provoking a write operation, writing data to the at least one log-structured storage device and writing redundant data to at least one fast storage device.

19. The method of claim 18, wherein the at least one log-structured storage device is divided into a plurality of extents for storing data, and the step of writing data comprises appending data to a current extent of the plurality of extents.

20. The method of claim 18, further comprising steps of:

determining whether the request provokes a read operation on the storage system; and

determining whether the at least one fast storage device is busy in response to the request provoking a read operation;

reading data from the at least one fast storage device in response to the at least one fast storage device not being busy; and

reading data from any one of the at least one log-structured storage device and the at least one fast storage device in response to the at least one fast storage device being busy.

21. The method of claim 20, wherein the step of determining whether the request provokes a read operation on the storage system further comprises steps of:

determining whether requested data is in a cache in response to the request provoking a read operation; and

reading the requested data from the cache in response to the requested data being in the cache.

22. The method of claim 18, further comprising writing data to a cache in response to the request provoking a write operation.

23. A method of recovering from a failed device in a storage system having at least one log-structured storage device and at least one fast storage device storing data redundant to data stored in the at least one log-structured storage device, the method comprising steps of:

determining whether the at least one log-structured storage device failed;

reading data from the at least one fast storage device in response to the at least one log-structured storage device being failed, the read data including data redundant to the data stored on the at least one failed log-structured storage device; and

writing the redundant data to a log-structured storage device that is operative in the storage system.

24. The method of claim 23, wherein the step of writing the redundant data comprises appending the redundant data to a current extent in the operative log-structured storage device.

25. The method of claim 23, further comprising steps of:

determining whether the at least one fast storage device failed;

reading data from the at least one log-structured storage device in response to the at least one fast storage device being failed, the data being read including data redundant to the data stored on the at least one failed fast storage device; and

writing the data redundant to the data stored on the failed fast storage device to an operative fast storage device in the storage system.

26. The method of claim 25, wherein the step of reading data from the at least one log-structured storage device comprises reading the redundant data from the at least one log-structured storage device in an order that minimizes positioning delays in the at least one log-structured storage device.

27. A computer storage apparatus comprising:

a controller means for performing data operations in a storage array means for storing data; and

the storage array means comprises a first sub-array means for storing data using a log-structured algorithm and a fast storage sub-array means for storing data redundant to data stored in the first sub-array means, wherein the fast storage sub-array means includes one or more fast storage devices.

28. The computer storage apparatus of claim 27, wherein the first sub-array includes at least one log-structured storage device divided into extents and the controller means is operable to use the log-structured algorithm to write data into at least one of the extents.

29. The computer storage apparatus of claim 28, further comprising a memory means for storing metadata associated with the data written into the at least one extent using the log-structured algorithm.

30. The computer storage apparatus of claim 28, wherein the controller means is operable to clean an extent of the at least one log-structured storage device, the extent being cleaned including valid data.

31. The computer storage apparatus of claim 30, wherein cleaning the extent comprises reading data from the fast storage means, the read data corresponding to valid data in the extent being cleaned;

and writing the valid data into at least one extent in the first sub-array means using the log-structured algorithm.

32. The computer storage apparatus of claim 31, wherein the controller means is operable to mark the extent being cleaned as clean after all the valid data is appended to the at least one extent.

33. The computer storage apparatus of claim 27, wherein the controller means attempts to perform a read operation on the fast storage sub-array means before attempting to perform the read operation on the first sub-array means.

34. The computer storage apparatus of claim 27, wherein the controller means is operable to write data to the first sub-array means and write the redundant data to the fast storage sub-array means.