File mode RAID subsystem

• US 20050021615A1
• Filed: 08/17/2004
• Published: 01/27/2005
• Est. Priority Date: 12/06/2001
• Status: Active Grant

• Current Assignee: Avago Technologies International Sales Pte Limited (Broadcom, Inc.)
• Original Assignee: Broadcom Corporation (Broadcom, Inc.), RAIDCore, Inc. (Broadcom, Inc.)
• Inventors: Arnott, Randy M., Noya, Eric S.
• Granted Patent: US 7,054,998 B2
• Time in Patent Office: Days
• US Class Current: 709/203
• CPC Class Codes: G06F 16/10 File systems; File servers
• CPC Class Codes: G06F 2003/0697 device management, e.g. handlers, drivers, I/O schedulers
• CPC Class Codes: G06F 3/0601 Interfaces specially adapted for storage systems

Abstract

A method and system enables data redundancy across servers, networks, and controllers by using standard redundant files as underlying storage for RAID subsystem configurations. A redundant array of independent disk (RAID) subsystem includes a front-end interface configured to process non-redundant requests received from a primary file system communicating with an application program. A back-end interface of the RAID subsystem is configured to process redundant requests corresponding to the non-redundant requests. The redundant requests to be issued to a secondary file system communicates with a block mode device including multiple physical storage devices.

Claims

1. A method for accessing stored data maintained by a redundant array of independent disk (RAID) subsystem, comprising:

generating, in an application program, a non-redundant file I/O request to a primary file system;

receiving, in a RAID subsystem, the non-redundant block I/O request from the primary file system;

translating, in the RAID subsystem, the non-redundant block I/O request into a redundant file I/O request for redundant file data maintained by the RAID subsystem; and

translating, in a secondary file system, the redundant file I/O request into a plurality of non-redundant block I/O requests for a block mode device storing the redundant file data.

2. The method of claim 1 further comprising:

accessing non-redundant block I/O data in the block mode device in response to receiving the plurality of non-redundant block I/O request;

accessing redundant file I/O data in the secondary file system in response to the redundant file I/O request; and

accessing non-redundant block I/O data in the primary file system in response to the non-redundant block I/O requests.

3. The method of claim 1 wherein the block mode device includes a plurality of disk drives, and further comprising:

storing a particular non-redundant block on at least at least two disk drives of the block mode device.

4. The method of claim 1 wherein the primary file system includes a primary client side and a primary server side and the secondary file system includes a secondary client side and a secondary server side, and further comprising:

communicating between the primary client side and the primary server side of the primary file system via a first network; and

communicating between the secondary client side and the secondary server side of the secondary file system via a second network.

5. The method of claim 1 wherein the first network is a local area network, and the second network is a wide area network.

6. The method of claim 1 further comprising:

maintaining at least two redundant files via the secondary file system for each single file maintained by the primary file system.

7. The method of claim 1 wherein the RAID subsystem maintains the redundant files using the secondary file system, each redundant file comprising a collection of related blocks including metadata and user data distributed over multiple physical devices of the block mode device such that each redundant file can be recovered after a failure of any one of the multiple physical devices.

8. The method of claim 1 wherein the secondary file system includes a server secondary file system and a local secondary file system, and further comprising:

executing the server secondary file system on a server computer; and

executing the local secondary file system on a client computer.

9. The method of claim 3 wherein the plurality of disk drives are connected to multiple different computers.

10. The method of claim 9 wherein the multiple different computers are connected by a wide area network.

11. The method of claim 1 further comprising at least one client computer and at least one server computer, and further comprising:

executing the RAID subsystem on the at least one client computer.

12. The method of claim 1 further comprising at least one client computer and at least one server computer, and further comprising:

executing the RAID subsystem on the at least one server computer.

13. A redundant array of independent disk (RAID) subsystem, comprising:

a front-end interface configured to process non-redundant I/O requests received from a primary file system communicating with an application program; and

a back-end interface configured to process redundant I/O requests corresponding to the non-redundant I/O requests, the redundant I/O requests to be issued to a secondary file system communicating with a block mode device.

14. The RAID subsystem of claim 13 wherein the non-redundant I/O request are block mode requests, and the redundant requests I/O are file mode requests.

15. The RAID system of claim 13 wherein the block mode device includes a plurality of disk drives, at least two disk drives storing a different copy of identical non-redundant block I/O data for a particular non-redundant block I/O request.

17. The RAID system of claim 13 further comprising:

a first network connecting a client side with a server side of the primary file system; and

a second network connecting a client side with a server side of the secondary file system.

18. The RAID subsystem of claim 13 further comprising:

at least one client computer and at least one server computer, and further comprising;

executing the RAID subsystem on the at least one client computer.

19. The RAID subsystem of claim 13 further comprising:

at least one client computer and at least one server computer, and further comprising;

executing the RAID subsystem on the at least one server computer.

16. The RAID subsystem 15 wherein the plurality of disk drives are connected multiple different computers and the multiple different computers are connected via a wide area network.

20. A redundant array of independent disk (RAID) subsystem, comprising:

a front-end interface configured to process non-redundant block I/O requests received from an application program using non-redundant block mode access; and

a back-end interface configured to process redundant file I/O requests corresponding to the non-redundant requests, the redundant requests to be issues to a secondary file system using redundant file mode access, the secondary file system communicating with a block mode device storing redundant files.