Dynamically controlling temporary compromise on data redundancy

• US 9,513,820 B1
• Filed: 04/07/2014
• Issued: 12/06/2016
• Est. Priority Date: 04/07/2014
• Status: Active Grant

• Current Assignee: Pure Storage, Inc.
• Original Assignee: Pure Storage, Inc.
• Inventors: Shalev, Ori
• Primary Examiner(s): Rutz, Jared
• Assistant Examiner(s): Baughman, William E
• Application Number:
  US14/246,900
• Time in Patent Office: 974 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06F 11/1076 Parity data used in redundant arrays of independent storages, e.g. in RAID systems
• CPC Class Codes: G06F 3/0611 in relation to response time
• CPC Class Codes: G06F 3/0614 Improving the reliability of storage systems
• CPC Class Codes: G06F 3/0644 Management of space entities, e.g. partitions, extents, pools
• 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

Abstract

Systems and methods for determining when to allow a temporary compromise on redundancy in a storage system. When servicing write requests, the storage system may utilize data redundancy techniques when writing data to backend storage devices. The operating conditions of the storage system are tracked and early acknowledgements for write requests may be permitted when the storage system is healthy enough. If the number of unacknowledged writes is greater than a programmable threshold, then early acknowledges may be prohibited. Also, if the number of ongoing rebuild processes is greater than a programmable threshold, then early acknowledges may be prohibited.

Claims

1. A method comprising:

receiving a write request from a client of a storage system, the storage system comprising a plurality of storage devices;

determining that a first condition is satisfied, wherein determining that the first condition is satisfied includes;

detecting that a number of unacknowledged writes of data to storage devices of the plurality of storage devices is below a first threshold; and

detecting that a number of ongoing rebuild processes within the storage system is below a second threshold, wherein the second threshold is different from the first threshold; and

in response to determining that the first condition is satisfied, issuing an early acknowledgement to the client for the write request,

wherein issuing the early acknowledgement comprises sending an acknowledgement of the write request to the client prior to completing all writes of data associated with the write request to the storage system.

2. The method as recited in claim 1, wherein the first threshold and the second threshold are programmable.

3. The method as recited in claim 1 wherein determining that the first condition is satisfied further includes detecting that a weighted sum of the number of unacknowledged writes of data to storage devices and the number of ongoing rebuild processes within the storage system is below a third threshold.

4. The method as recited in claim 1, further comprising:

tracking a performance history of one or more devices of the plurality of storage devices; and

wherein issuing an early acknowledgement to the client for the write request is based at least in part on said performance history.

5. A system comprising:

a plurality of storage devices; and

a storage controller coupled to the plurality of storage devices;

wherein the storage controller is configured to;

receive a write request from a client of a storage system, the storage system comprising a plurality of storage devices;

determine that a first condition is satisfied, wherein determining that the first condition is satisfied includes;

detecting that a number of unacknowledged writes of data to storage devices of the plurality of storage devices is below a first threshold; and

detecting that a number of ongoing rebuild processes within the storage system is below a second threshold, wherein the second threshold is different from the first threshold; and

in response to determining that the first condition is satisfied, issue an early acknowledgement to the client for the write request,

wherein issuing the early acknowledgement comprises sending an acknowledgement of the write request to the client prior to completing all writes of data associated with the write request to the storage system.

6. The system as recited in claim 5, wherein the threshold and the second threshold are programmable.

7. The system as recited in claim 5 wherein determining that the first condition is satisfied further includes detecting that a weighted sum of the number of unacknowledged writes of data to storage devices and the number of ongoing rebuild processes within the storage system is below a third threshold.

8. The system as recited in claim 5, wherein the storage controller is further configured to:

track a performance history of the plurality of storage devices; and

wherein issuing an early acknowledgement to the client for the write request is based at least in part on said performance history.

9. A non-transitory computer readable storage medium storing program instructions, wherein the program instructions are executable by a processor to:

receive a write request from a client of a storage system, the storage system comprising a plurality of storage devices;

determine that a first condition is satisfied, wherein determining that the first condition is satisfied includes;

detecting that a number of unacknowledged writes of data to storage devices of the plurality of storage devices is below a first threshold; and

detecting that a number of ongoing rebuild processes within the storage system is below a second threshold, wherein the second threshold is different from the first threshold; and

in response to determining that the first condition is satisfied, issue an early acknowledgement to the client for the write request,

wherein issuing the early acknowledgement comprises sending an acknowledgement of the write request to the client prior to completing all writes of the data associated with the write request to the storage system.

10. The non-transitory computer readable storage medium as recited in claim 9, wherein the threshold and the second threshold are programmable.

11. The non-transitory computer readable storage medium as recited in claim 9 wherein determining that the first condition is satisfied further includes detecting that a weighted sum of the number of unacknowledged writes of data to storage devices and the number of ongoing rebuild processes within the storage system is below a third threshold.