Adaptive RAID for an SSD environment

• US 8,775,868 B2
• Filed: 09/28/2010
• Issued: 07/08/2014
• Est. Priority Date: 09/28/2010
• Status: Active Grant

• Current Assignee: Pure Storage, Inc.
• Original Assignee: Pure Storage, Inc.
• Inventors: Colgrove, John, Hayes, John, Hong, Bo, Miller, Ethan
• Primary Examiner(s): SCHELL, JOSEPH O
• Application Number:
  US12/892,894
• Publication Number:
  US 20120079318A1
• Time in Patent Office: 1,379 Days
• Field of Search: 714/6.22, 714/6.23, 714/6.24, 714/6.32, 714/6.1, 714/6.2, 714/6.21, 711/114
• US Class Current: 714/6.24
• CPC Class Codes: G06F 11/1008 in individual solid state devices G06F11/1004 takes precedence
• CPC Class Codes: G06F 11/1068 in sector programmable memories, e.g. flash disk G06F11/1072 takes precedence
• CPC Class Codes: G06F 11/1076 Parity data used in redundant arrays of independent storages, e.g. in RAID systems
• CPC Class Codes: G06F 11/108 Parity data distribution in semiconductor storages, e.g. in SSD
• CPC Class Codes: G06F 11/2053 where persistent mass storage functionality or persistent mass storage control functionality is redundant error detection or correction in information storage based on relative movement between record carrier and transducer G11B20/18
• CPC Class Codes: G06F 3/0619 in relation to data integrity, e.g. data losses, bit errors
• 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 3/0665 at area level, e.g. provisioning of virtual or logical volumes
• CPC Class Codes: G06F 3/0688 Non-volatile semiconductor memory arrays
• CPC Class Codes: G06F 3/0689 Disk arrays, e.g. RAID, JBOD

Abstract

A system and method for adaptive RAID geometries. A computer system comprises client computers and data storage arrays coupled to one another via a network. A data storage array utilizes solid-state drives and Flash memory cells for data storage. A storage controller within a data storage array is configured to determine a first RAID layout for use in storing data, and write a first RAID stripe to the device group according to the first RAID layout. In response to detecting a first condition, the controller is configured to determine a second RAID layout which is different from the first RAID layout, and write a second RAID stripe to the device group according to the second layout, whereby the device group concurrently stores data according to both the first RAID layout and the second RAID layout.

Claims

1. A computer system comprising:

a client computer configured to send read and write requests over a network; and

a data storage subsystem coupled to the network configured to receive the read and write requests, wherein the subsystem comprises a plurality of data storage locations on a device group including a plurality of storage devices;

wherein the data storage subsystem further comprises a storage controller configured to;

determine a first RAID layout and second RAID layout for use in storing data, wherein the second RAID layout is different from the first RAID layout;

write a first RAID stripe to a first subgroup of devices of the plurality of storage devices according to the first RAID layout; and

write a second RAID stripe to a second subgroup of devices of the plurality of storage devices according to the second layout;

write a third RAID stripe to a third subgroup of devices of the plurality of storage devices according to the second layout, wherein the third subgroup of devices includes one or more devices not included in the second subgroup;

whereby the device group concurrently stores data according to both the first RAID layout and the second RAID layout;

wherein in response to detecting a change in reliability of a given device of the first subgroup of devices, the storage controller is configured to;

maintain a first amount of space in a given device for storing intra-device redundancy data for a given RAID stripe; and

concurrently maintain a second amount of space in a device other than the given device for storing intra-device redundancy data for the given RAID stripe, wherein the second amount is different than the first amount.

2. The computer system as recited in claim 1, wherein the storage controller is further configured to detect characteristics which affect reliability, wherein the characteristics include one or more of an age of data stored in one or more of the plurality of devices, an amount of free space in one or more of the plurality of storage devices, an expected life of data to be written in the space allocated, an expected life of data currently stored in one or more of the plurality of storage devices, a device age, an error rate, a number of errors, a number of recoverable errors, a number of unrecoverable errors, and an allocation state.

3. The computer system as recited in claim 1, wherein the storage controller is further configured to shred protection data in the plurality of storage devices, whereby the storage controller is configured to at least:

determine that a level of protection for given data is to be changed;

identify particular storage locations storing protection data corresponding to the given data; and

de-allocate the particular storage locations.

4. The computer system as recited in claim 3, wherein the storage controller is further configured to reallocate one or more of the particular storage locations for storing non-protection data during a subsequent write, wherein the given data is not copied to a new location.

5. The computer system as recited in claim 4, wherein a reallocated particular storage location need not use a same size or alignment for the non-protection data as was used for the protection data.

6. The computer system as recited in claim 1, wherein a storage device of the plurality of storage devices is configured to erase data in erase block size units, and wherein in response to detecting an access to a portion of data smaller in size than an erase block on the storage device has failed but other accesses to data on the storage device succeed, the storage controller is configured to:

identify a particular erase block sized portion of data that includes the portion of data corresponding to the access that failed;

rebuild the portion of data to form a rebuilt portion of data; and

cause the storage device to write the particular erase block sized portion of data including the rebuilt portion of data.

7. The computer system as recited in claim 1, wherein in response to detecting an access to a portion of data has failed, the storage controller is configured to:

rebuild the portion of data corresponding to the access that failed to form a rebuilt portion of data;

shred protection data in a particular storage location of the plurality of storage devices; and

store the rebuilt portion of data in the particular storage location.

8. The computer system as recited in claim 1, wherein the storage controller is further configured to issue a command to a storage device of the plurality of storage devices which causes the storage device to remap and rewrite a portion of data stored in the storage device.

9. A method for use in a computer system, the method comprising:

receiving read and write requests at a data storage subsystem, wherein the subsystem comprises a plurality of data storage locations on a device group including a plurality of storage devices;

determining a first RAID layout for use in storing data;

writing a first RAID stripe to a first subgroup of devices of the plurality of storage devices according to the first RAID layout;

determining a second RAID layout which is different from the first RAID layout;

writing a second RAID stripe to a second subgroup of devices of the plurality of storage devices according to the second layout;

writing a third RAID stripe to a third subgroup of devices of the plurality of storage devices according to the second layout, wherein the third subgroup of devices includes one or more devices not included in the second subgroup;

whereby the device group concurrently stores data according to both the first RAID layout and the second RAID layout; and

writing a fourth RAID stripe to the plurality of storage devices, wherein an amount of intra-device redundancy data maintained on a first device for the fourth RAID stripe is different from an amount of intra-device redundancy data concurrently maintained on a second device for the fourth RAID stripe.

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

(i) the first RAID layout, (ii) the second RAID layout, and (iii) both a first subgroup and a second subgroup of devices within the device group to store data corresponding to the first RAID layout and the second RAID layout, respectively, wherein the second subgroup is different from the first subgroup.

11. The method as recited in claim 9, wherein characteristics which affect reliability include one or more of an age of data stored in one or more of the plurality of devices, an amount of free space in one or more of the plurality of storage devices, an expected life of data to be written in the space allocated, an expected life of data currently stored in one or more of the plurality of storage devices, a device age, an error rate, a number of errors, a number of recoverable errors, a number of unrecoverable errors, and an allocation state.

12. The method as recited in claim 9, further comprising shredding protection data in the plurality of storage devices, whereby the method comprises:

determining that a level of protection for given data is to be changed;

identifying particular storage locations storing protection data corresponding to the given data; and

de-allocating the particular storage locations.

13. The method as recited in claim 12, reallocating one or more of the particular storage locations for storing non-protection data during a subsequent write, wherein the given data is not copied to a new location.

14. The method as recited in claim 13, wherein a reallocated particular storage location need not use a same size or alignment for the non-protection data as was used for the protection data.

15. The method as recited in claim 9, wherein a storage device of the plurality of storage devices is configured to erase data in erase block size units, and wherein in response to detecting an access to a portion of data smaller in size than an erase block on the storage device has failed but other accesses to data on the storage device succeed, the method further comprises:

identifying a particular erase block sized portion of data that includes the portion of data corresponding to the access that failed;

rebuilding the portion of data to form a rebuilt portion of data; and

causing the storage device to write the particular erase block sized portion of data including the rebuilt portion of data.

16. A non-transitory computer readable storage medium storing program instructions, wherein the program instructions are executable to:

receive read and write requests at a data storage subsystem, wherein the subsystem comprises a plurality of data storage locations on a device group including a plurality of storage devices;

determine a first RAID layout and second RAID layout for use in storing data, wherein the second RAID layout is different from the first RAID layout;

write a first RAID stripe to a first subgroup of devices of the plurality of storage devices according to the first RAID layout; and

write a second RAID stripe to a second subgroup of devices of the plurality of storage devices according to the second layout;

write a third RAID stripe to a third subgroup of devices of the plurality of storage devices according to the second layout, wherein the third subgroup of devices includes one or more devices not included in the second subgroup;

write data according to a given layout, wherein writing said data according to the given layout comprises selecting from any of the plurality of devices for storing the data without regard to devices used in a previous write of data according to the given layout;

wherein in response to detecting a change in reliability of a given device of the first subgroup of devices, the program instructions are further executable to;

maintain a first amount of space in a given device for storing intra-device redundancy data for a given RAID stripe; and

concurrently maintain a second amount of space in a device other than the given device for storing intra-device redundancy data for the given RAID stripe, wherein the second amount is different than the first amount.