Performance improvements in a storage system

• US 9,773,007 B1
• Filed: 12/01/2014
• Issued: 09/26/2017
• Est. Priority Date: 12/01/2014
• Status: Expired due to Fees

• Current Assignee: Pure Storage, Inc.
• Original Assignee: Pure Storage, Inc.
• Inventors: Golden, Christopher, Vachharajani, Neil, Grunwald, David, Smith, Scott, Cao, Jianting, Paulsen, Luke, Gonzalez, Ricardo
• Primary Examiner(s): Ly, Anh
• Application Number:
  US14/556,611
• Time in Patent Office: 1,030 Days
• CPC Class Codes: G06F 12/08 in hierarchically structured memory systems, e.g. virtual memory systems
• CPC Class Codes: G06F 16/128 Details of file system snapshots on the file-level, e.g. snapshot creation, administration, deletion use of snapshots for error detection or correction G06F11/14, G06F11/16
• CPC Class Codes: G06F 3/0608 Saving storage space on storage systems
• CPC Class Codes: G06F 3/0646 Horizontal data movement in storage systems, i.e. moving data in between storage devices or systems
• CPC Class Codes: G06F 3/065 Replication mechanisms
• CPC Class Codes: G06F 3/0652 Erasing, e.g. deleting, data cleaning, moving of data to a wastebasket
• CPC Class Codes: G06F 3/0688 Non-volatile semiconductor memory arrays
• CPC Class Codes: G06F 3/0689 Disk arrays, e.g. RAID, JBOD

Abstract

Systems, methods, and computer readable storage mediums for optimizing performance in a storage system. A volume may be mapped to an anchor medium through an indirection layer. When a snapshot is taken of the volume, the allocation of a new anchor medium is delayed until a write operation targeting the volume is received. When a write operation targeting the volume is received, a new anchor medium may be allocated for only a portion of the volume corresponding to the portion of the volume which is targeted by the write operation. Also, when a snapshot is taken of the volume, read optimization operations are performed only the portions of the volume which have changed since the previous snapshot.

Claims

1. A system comprising:

one or more storage devices; and

a storage controller coupled to the one or more storage devices, wherein the storage controller is configured to;

allocate an indirection layer for a volume, wherein the indirection layer is mapped directly below the volume and above a first anchor medium, wherein the first anchor medium is a logical grouping of data;

wherein the indirection layer is a link between the volume and the first anchor medium of the volume;

map read and write operations targeting the volume through the indirection layer;

in response to detecting a request to perform a first snapshot of the volume, convert a first anchor medium from a read-write state to a read-only state;

in response to detecting a write operation targeting the volume and subsequent to detecting the request to perform the first snapshot of the volume, allocate a second anchor medium that maps directly below the indirection layer and above the first anchor medium; and

perform the write operation to the second anchor medium.

2. The system as recited in claim 1, wherein a size of the second anchor medium is based at least in part on one or more of a size of a write operation targeted to the second anchor medium, a historical write pattern, a size of a data structure used to identify and/or maintain medium mappings, and fragmentation of the data structure.

3. The system as recited in claim 2, wherein responsive to detecting that a number of levels in a medium hierarchy of the volume has exceeded a threshold, the storage controller is configured to perform read optimization operations on the volume to reduce the number of levels in the medium hierarchy of the volume, and wherein responsive to determining a size of a medium mapping table has exceeded a given threshold, the storage controller is configured to reduce the threshold for the number of levels in the medium hierarchy of the volume.

4. The system as recited in claim 1, wherein responsive to detecting a request to perform a second snapshot of the volume, the storage controller is configured to:

perform read optimization operations on the second anchor medium; and

prevent read optimization operations from being performed on portions of the volume corresponding to the first anchor medium.

5. The system as recited in claim 1, wherein the storage controller is further configured to:

detect a request to copy at least a first portion of the volume as part of a first extended copy operation;

convert a first portion of a first anchor medium corresponding to the first portion of the volume to a read-only state, responsive to detecting the request to perform the first extended copy operation; and

delay allocation of a second anchor medium until a write operation targeting the first portion of the volume is received.

6. A method comprising:

allocating an indirection layer for a volume, wherein the indirection layer is mapped directly below the volume and above a first anchor medium, wherein the first anchor medium is a logical grouping of data;

wherein the indirection layer is a link between the volume and the first anchor medium of the volume;

mapping read and write operations targeting the volume through the indirection layer;

in response to detecting a request to perform a first snapshot of the volume, converting a first anchor medium from a read-write state to a read-only state;

in response to detecting a write operation targeting the volume and subsequent to detecting the request to perform the first snapshot of the volume, allocating a second anchor medium that maps directly below the indirection layer and above the first anchor medium; and

performing the write operation to the second anchor medium.

7. The method as recited in claim 6, wherein a size of the second anchor medium is based at least in part on one or more of a size of a write operation targeted to the second anchor medium, a historical write pattern, a size of a data structure used to identify and/or maintain medium mappings, and fragmentation of the data structure.

8. The method as recited in claim 7, wherein responsive to detecting that a number of levels in a medium hierarchy of the volume has exceeded a threshold, the method further comprising performing read optimization operations on the volume to reduce the number of levels in the medium hierarchy of the volume, and wherein responsive to determining a size of a medium mapping table has exceeded a given threshold, the method further comprising reducing the threshold for the number of levels in the medium hierarchy of the volume.

9. The method as recited in claim 6, further comprising:

performing read optimization operations only on the second anchor medium; and

preventing read optimization operation from being performed on portions of the volume corresponding to the first anchor medium.

10. The method as recited in claim 6, further comprising:

detecting a request to copy at least a first portion of the volume as part of a first extended copy operation;

converting a first portion of a first anchor medium corresponding to the first portion of the volume to a read-only state, responsive to detecting the request to perform the first extended copy operation; and

delaying allocation of a second anchor medium until a write operation targeting the first portion of the volume is received.

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

allocate an indirection layer for a volume, wherein the indirection layer is mapped directly below the volume and above a first anchor medium, wherein the first anchor medium is a logical grouping of data;

wherein the indirection layer is a link between the volume and the first anchor medium of the volume;

map read and write operations targeting the volume through the indirection layer;

in response to detecting a request to perform a first snapshot of the volume, convert a first anchor medium from a read-write state to a read-only state;

in response to detecting a write operation targeting the volume and subsequent to detecting the request to perform the first snapshot of the volume, allocate a second anchor medium that maps directly below the indirection layer and above the first anchor medium; and

perform the write operation to the second anchor medium.

12. The non-transitory computer readable storage medium as recited in claim 11, wherein a size of the second anchor medium is based at least in part on one or more of a size of a write operation targeted to the second anchor medium, a historical write pattern, a size of a data structure used to identify and/or maintain medium mappings, and fragmentation of the data structure.

13. The non-transitory computer readable storage medium as recited in claim 12, wherein responsive to detecting that a number of levels in a medium hierarchy of the volume has exceeded a threshold, the program instructions are further executable by a processor to perform read optimization operations on the volume to reduce the number of levels in the medium hierarchy of the volume, and wherein responsive to determining a size of a medium mapping table has exceeded a given threshold, the program instructions are further executable by a processor to reduce the threshold for the number of levels in the medium hierarchy of the volume.

14. The non-transitory computer readable storage medium as recited in claim 11, wherein the program instructions are further executable by a processor to:

detect a request to copy at least a first portion of the volume as part of a first extended copy operation;

convert a first portion of a first anchor medium corresponding to the first portion of the volume to a read-only state, responsive to detecting the request to perform the first extended copy operation; and

delay allocation of a second anchor medium until a write operation targeting the first portion of the volume is received.