Method and apparatus for decomposing I/O tasks in a RAID system

• US 7,926,059 B2
• Filed: 05/13/2009
• Issued: 04/12/2011
• Est. Priority Date: 03/21/2002
• Status: Expired due to Fees

• Current Assignee: NetApp, Inc.
• Original Assignee: NetApp, Inc.
• Inventors: Viswanathan, Srinivasan, Strange, Stephen H., Schoenthal, Scott, Sundaram, Rajesh, Leong, James, Doucette, Douglas P.
• Primary Examiner(s): WU, QING YUAN
• Application Number:
  US12/465,179
• Publication Number:
  US 20090222829A1
• Time in Patent Office: 699 Days
• Field of Search: 718/102, 711/100, 711/114
• US Class Current: 718/102
• CPC Class Codes: G06F 11/2087 with a common controller
• CPC Class Codes: G06F 3/0601 Interfaces specially adapted for storage systems
• CPC Class Codes: G06F 3/061 Improving I/O performance
• CPC Class Codes: G06F 3/0611 in relation to response time
• CPC Class Codes: G06F 3/0655 Vertical data movement, i.e. input-output transfer; data movement between one or more hosts and one or more storage devices
• CPC Class Codes: G06F 3/0659 Command handling arrangements, e.g. command buffers, queues, command scheduling
• CPC Class Codes: G06F 3/0689 Disk arrays, e.g. RAID, JBOD
• CPC Class Codes: G06F 9/4881 Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues

Abstract

A data access request to a file system is decomposed into a plurality of lower-level I/O tasks. A logical combination of physical storage components is represented as a hierarchical set of objects. A parent I/O task is generated from a first object in response to the data access request. A child I/O task is generated from a second object to implement a portion of the parent I/O task. The parent I/O task is suspended until the child I/O task completes. The child I/O task is executed in response to an occurrence of an event that a resource required by the child I/O task is available. The parent I/O task is resumed upon an event indicating completion of the child I/O task. Scheduling of any child I/O task is not conditional on execution of the parent I/O task, and a state diagram regulates the child I/O tasks.

Claims

1. A method for decomposing a data access request from a storage operating system of a storage system into a plurality of lower-level I/O tasks, comprising:

representing a logical combination and configuration of one or more physical storage components of the storage operating system as a hierarchical set of objects;

generating a parent I/O task from a first object in the hierarchical set of objects in response to the data access request from the storage operating system;

generating a child I/O task from a second object in the hierarchical set of objects to implement at least a portion of the parent I/O task;

suspending the parent I/O task until the child I/O task is completed;

executing the child I/O task in response to a first event indicating a resource of the storage system required by the child I/O task is available, such that scheduling of the child I/O task is not conditional on execution of the parent I/O task; and

resuming the parent I/O task in response to a second event indicating to the parent I/O task the completion of the child I/O task.

2. The method of claim 1, further comprising:

using a storage disk as at least one of the one or more physical storage components.

3. The method of claim 1, further comprising:

using a plurality of storage disks as the one or more physical storage components.

4. The method of claim 1, further comprising:

using a Redundant Array of Independent Disks (RAID) as the one or more physical storage components.

5. The method of claim 1, further comprising:

using non-volatile RAM (NVRAM) as at least one of the one or more physical storage components.

6. The method of claim 1, further comprising:

using a memory as at least one of the one or more physical storage components.

7. The method of claim 1, further comprising:

using a writeable persistent storage device as at least one of the one or more physical storage components.

8. The method of claim 1, further comprising:

using a tape as at least one of the one or more physical storage components.

9. The method of claim 1, further comprising:

scheduling for execution a grandchild I/O task in response to a third event such that execution of the grandchild I/O task is not conditional on execution of at least of the parent I/O task and the child I/O task; and

indicating to the child I/O task from the grandchild I/O task that the grandchild I/O task has completed execution.

10. The method of claim 1, wherein the storage operating system comprises a file system.

11. The method of claim 1, further comprising:

queuing the parent I/O task and the child I/O task on separate I/O threads.

12. The method of claim 1, further comprising:

queuing the parent I/O task and the child I/O task on separate devices of the storage system.

13. The method of claim 1, further comprising:

queuing the parent I/O task and the child I/O task on separate processes executed at the storage system.

14. The method of claim 1, further comprising:

in response to generating the child I/O task, dynamically allocating the resource for the child I/O task.

15. The method of claim 1, further comprising:

in response to the second event, changing the child I/O task from a first state to a second state.

16. An apparatus, comprising:

a network adapter configured to receive a data access request;

a processor configured to represent a logical combination of one or more physical storage components of a storage operating system as a hierarchical set of objects;

the processor further configured to generate a parent I/O task from a first object in the hierarchical set of objects in response to the data access request;

the processor further configured to generate a child I/O task from a second object in the hierarchical set of objects to implement at least a portion of the parent I/O task;

the processor further configured to suspend the parent I/O task until the child I/O task is completed;

the processor further configured to execute the child I/O task in response to a first event indicating a resource required by the child I/O task is available, such that scheduling of the child I/O task is not conditional on execution of the parent I/O task; and

the processor further configured to resume the parent I/O task in response to a second event indicating to the parent I/O task the completion of the child I/O task.

17. The apparatus of claim 16, further comprising:

a storage disk used as at least one of the one or more physical storage components.

18. The apparatus of claim 16, further comprising:

a plurality of storage disks used as the one or more physical storage components.

19. The apparatus of claim 16, further comprising:

a Redundant Array of Independent Disks (RAID) used as the one or more physical storage components.

20. The apparatus of claim 16, further comprising:

non-volatile RAM (NVRAM) used as at least one of the one or more physical storage components.

21. The apparatus of claim 16, further comprising:

a memory used as at least one of the one or more physical storage components.

22. The apparatus of claim 16, further comprising:

a writeable persistent storage device used as at least one of the one or more physical storage components.

23. The apparatus of claim 16, further comprising:

a tape used as at least one of the one or more physical storage components.

24. The apparatus of claim 16, further comprising:

the processor further configured to schedule execution for a grandchild I/O task in response to a third event such that execution of the grandchild I/O task is not conditional on execution of the parent I/O task or the child I/O task; and

the processor further configured to indicate to the child I/O task that the grandchild I/O task has completed execution.

25. The apparatus of claim 16, wherein the storage operating system comprises a file system.

26. The apparatus of claim 16, wherein the parent I/O task and the child I/O task are queued on separate I/O threads.

27. The apparatus of claim 16, wherein the parent I/O task and the child I/O task are queued on separate devices.

28. The apparatus of claim 16, wherein the parent I/O task and the child I/O task are queued on separate processes.

29. The apparatus of claim 16, further comprising:

the processor further configured to, in response to generating the child I/O task, dynamically allocate the resource for the child I/O task.

30. The apparatus of claim 16, further comprising:

the processor further configured to, in response to the second event, change the child I/O task from a first state to a second state.

31. The apparatus of claim 16, wherein the apparatus comprises a storage system.

32. A computer-readable storage media, comprising:

the computer-readable storage media comprising instructions for execution by a processor of a storage system for a method of decomposing a data access request from a storage operating system of the storage system into a plurality of lower-level I/O tasks, the computer-readable storage media further comprising instructions for,

representing a logical combination and configuration of one or more physical storage components of the storage operating system as a hierarchical set of objects;

generating a parent I/O task from a first object in the hierarchical set of objects in response to the data access request from the storage operating system;

generating a child I/O task from a second object in the hierarchical set of objects to implement at least a portion of the parent I/O task;

suspending the parent I/O task until the child I/O task is completed;

executing the child I/O task in response to a first event indicating a resource of the storage system required by the child I/O task is available, such that scheduling of the child I/O task is not conditional on execution of the parent I/O task; and

resuming the parent I/O task in response to a second event indicating to the parent I/O task the completion of the child I/O task.