Drive extent based end of life detection and proactive copying in a mapped RAID (redundant array of independent disks) data storage system

• US 10,459,814 B2
• Filed: 11/09/2017
• Issued: 10/29/2019
• Est. Priority Date: 10/31/2017
• Status: Active Grant

• Current Assignee: Emc IP Holding Company LLC (Dell Technologies Inc.)
• Original Assignee: Emc IP Holding Company LLC (Dell Technologies Inc.)
• Inventors: Gao, Hongpo, Kang, Jamin, Gao, Jian, Gong, Shaoqin, Sun, Ree
• Primary Examiner(s): Kudirka, Joseph R
• Application Number:
  US15/808,119
• Publication Number:
  US 20190129815A1
• Time in Patent Office: 719 Days
• Field of Search: 714 613, 714 622- 63, 714763, 714764
• CPC Class Codes: G06F 11/0727 in a storage system, e.g. in a DASD or network based storage system drivers for digital recording or reproducing units G06F3/06; circuits for error detection or correction within digital recording or reproducing units G11B20/18; for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS], H04L67/1097
• CPC Class Codes: G06F 11/1076 Parity data used in redundant arrays of independent storages, e.g. in RAID systems
• CPC Class Codes: G06F 11/1084 Degraded mode, e.g. caused by single or multiple storage removals or disk failures
• CPC Class Codes: G06F 11/2094 Redundant storage or storage space G06F11/2056 takes precedence
• CPC Class Codes: G06F 11/3034 where the computing system component is a storage system, e.g. DASD based or network based digital input from or digital output to record carriers G06F3/06; digital recording or reproducing G11B20/18; for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS], H04L67/1097
• CPC Class Codes: G06F 3/0616 in relation to life time, e.g. increasing Mean Time Between Failures [MTBF]
• CPC Class Codes: G06F 3/0619 in relation to data integrity, e.g. data losses, bit errors
• CPC Class Codes: G06F 3/0644 Management of space entities, e.g. partitions, extents, pools
• CPC Class Codes: G06F 3/0647 Migration mechanisms
• CPC Class Codes: G06F 3/065 Replication mechanisms
• CPC Class Codes: G06F 3/0689 Disk arrays, e.g. RAID, JBOD

Abstract

Mapped RAID (Redundant Array of Independent Disks) technology divides individual drives into multiple drive extents, allocates the drive extents to RAID extent entries in a RAID mapping table, and performs “end of life” detection and proactive copying of data between data storage drives on a per drive extent basis. A given drive extent is determined to be “end of life” when the ratio of soft media errors to total I/O operations for the drive extent exceeds a threshold error ratio. Data stored on the drive extent is then proactively copied to a newly allocated drive extent, the RAID mapping table is modified so that the data is subsequently accessed from the newly allocated drive extent, and the drive extent is excluded from being used again to store host data. As a result, the rate at which the drives experience soft media errors is slowed, lengthening their effective life.

Claims

1. A method of providing RAID (Redundant Array of Independent Disks) data protection for at least one storage object in a data storage system, wherein the data storage system includes at least one storage processor and an array of data storage drives communicably coupled to the storage processor, the method comprising:

generating a RAID mapping table, wherein the RAID mapping table contains a plurality of RAID extent entries, wherein each RAID extent entry contained in the RAID mapping table indicates a predetermined total number of drive extents that each persistently store host data written to a corresponding one of a plurality of RAID extents within a logical address space that is mapped to the at least one storage object, wherein each drive extent comprises a unique contiguous region of non-volatile data storage located on one of the data storage drives, and wherein each one of the data storage drives has multiple drive extents located thereon;

for each I/O operation directed to the storage object, performing a monitoring operation by the storage processor, wherein the monitoring operation includes;

i) incrementing a total I/O operations counter corresponding to a target drive extent to which that I/O operation is directed, wherein the total I/O operations counter corresponding to the target drive extent stores a total number of I/O operations that have been directed to the target drive extent,

ii) receiving, from a data storage drive within which the target drive extent is located, a completion status for that I/O operation, and

iii) in response to detecting that the received completion status for that I/O operation indicates that a soft media error occurred within the data storage drive while performing that I/O operation on the target drive extent;

a) incrementing a soft media error counter corresponding to the target drive extent, wherein the soft media error counter corresponding to the target drive extent stores a total number of soft media errors that have occurred while performing I/O operations on the target drive extent,

b) calculating an error ratio for the target drive extent, wherein the error ratio for the target drive extent comprises a ratio of a current value of the soft media error counter corresponding to the target drive extent to a current value of the total I/O operations counter corresponding to the target drive extent, and

c) in response to detecting that the error ratio for the target drive extent exceeds a threshold error ratio, performing a proactive copy operation on the target drive extent that copies all host data stored on the target drive extent to a newly allocated drive extent, wherein performing the proactive copy operation on the target drive extent also modifies a RAID extent entry in the RAID mapping table that stored an indication of the target drive extent to store an indication of the newly allocated drive extent, whereby the host data copied from the target drive extent to the newly allocated drive extent is accessed by subsequently received I/O operations on the newly allocated drive extent.

2. The method of claim 1, wherein performing the proactive copy operation on the target drive extent further includes setting a dead flag corresponding to the target drive extent, and wherein the method further comprises:

responsive to the dead flag corresponding to the target drive extent being set, preventing the target drive extent from subsequently being allocated to any RAID mapping table entry in the RAID mapping table.

3. The method of claim 2, further comprising:

receiving an end of life indication from the data storage drive; and

in response to receiving the end of life indication from the data storage drive, performing proactive copy operations on drive extents located within the data storage drive, wherein the proactive copy operations only copy host data stored on those drive extents located within the data storage drive that are indicated by particular RAID extent entries in the RAID mapping table to newly allocated drive extents, and wherein the proactive copy operations also modify the particular RAID extent entries in the RAID mapping table that stored indications of the drive extents located within the data storage drive to store indications of the newly allocated drive extents, whereby the host data copied from the drive extents located within the data storage drive that were indicated by RAID extent entries in the RAID mapping table to the newly allocated drive extents is subsequently accessed by subsequently received I/O operations on the newly allocated drive extents.

4. The method of claim 3, wherein performing proactive copy operations on all drive extents located within the data storage drive further includes setting a dead flag corresponding to each one of the drive extents located within the data storage drive, and wherein the method further comprises:

responsive to the dead flag corresponding to each one of the drive extents located within the data storage drive being set, preventing the drive extents located within the data storage drive from subsequently being allocated to any RAID mapping table entry in the RAID mapping table.

5. The method of claim 4, wherein performing the proactive copy operation on all drive extents located within the data storage drive further comprises:

setting an end of life flag corresponding to each one of the drive extents that are located within the data storage drive and that are also indicated by RAID extent entries in the RAID mapping table;

detecting, by a background process, the set end of life flags corresponding to the drive extents that are located within the data storage drive and that are also indicated by RAID extent entries in the RAID mapping table; and

in response to detecting, by the background process, the set end of life flags corresponding to the drive extents that are located within the data storage drive and that are also indicated by RAID extent entries in the RAID mapping table, copying, by the background process, all the host data stored on drive extents that are located within the data storage drive and that are also indicated by RAID extent entries in the RAID mapping table to the newly allocated drive extents, and modifying, by the background process, each one of the RAID extent entries in the RAID mapping table that stored an indication of one of the drive extents located within the data storage drive to instead store an indication of one of the newly allocated drive extents.

6. The method of claim 3, wherein the data storage drive within which the target drive extent is located includes a set of reserved disk sectors that are used to replace failed disk sectors;

and

wherein the received completion status for that I/O operation that indicates that a soft media error occurred within the data storage drive while performing that I/O operation on the target drive extent indicates that one of the set of reserved disk sectors was used to replace an original disk sector to which that I/O operation was directed within the target drive extent, in response to a failure of the original disk sector.

7. The method of claim 6, wherein the end of life indication received from the data storage drive indicates that a threshold maximum number of the reserved disk sectors have been used to replace original disk sectors to which I/O operations were directed, in response to failures of the original disk sectors.

8. The method of claim 7, wherein the threshold maximum number of the reserved disk sectors comprises the total number of reserved disk sectors in the data storage drive.

9. The method of claim 2, wherein performing the proactive copy operation on the target drive extent further comprises:

setting an end of life flag corresponding to the target drive extent;

detecting, by a background process, the set end of life flag corresponding to the target drive extent; and

in response to detecting the set end of life flag corresponding to the target drive extent, copying, by the background process, all the host data stored on the target drive extent to the newly allocated drive extent, and modifying, by the background process, the RAID extent entry in the RAID mapping table that stored an indication of the target drive extent to store an indication of the newly allocated drive extent.

10. A data storage system that provides RAID (Redundant Array of Independent Disks) data protection for a storage object, comprising:

at least one storage processor including processing circuitry and a memory;

an array of data storage drives communicably coupled to the at least one storage processor; and

wherein the memory has program code stored thereon, wherein the program code, when executed by the processing circuitry, causes the processing circuitry to;

generate a RAID mapping table, wherein the RAID mapping table contains a plurality of RAID extent entries, wherein each RAID extent entry contained in the RAID mapping table indicates a predetermined total number of drive extents that each persistently store host data written to a corresponding one of a plurality of RAID extents within a logical address space that is mapped to the storage object, wherein each drive extent comprises a unique contiguous region of non-volatile data storage located on one of the data storage drives, and wherein each one of the data storage drives has multiple drive extents located thereon;

for each I/O operation directed to the storage object, perform a monitoring operation by the storage processor, at least in part by causing the processing circuitry to;

i) increment a total I/O operations counter corresponding to a target drive extent to which that I/O operation is directed, wherein the total I/O operations counter corresponding to the target drive extent stores a total number of I/O operations that have been directed to the target drive extent,

ii) receive, from a data storage drive within which the target drive extent is located, a completion status for that I/O operation, and

iii) in response to detecting that the received completion status for that I/O operation indicates that a soft media error occurred within the data storage drive while performing that I/O operation on the target drive extent;

a) increment a soft media error counter corresponding to the target drive extent, wherein the soft media error counter corresponding to the target drive extent stores a total number of soft media errors that have occurred while performing I/O operations on the target drive extent,

b) calculate an error ratio for the target drive extent, wherein the error ratio for the target drive extent comprises a ratio of a current value of the soft media error counter corresponding to the target drive extent to a current value of the total I/O operations counter corresponding to the target drive extent, and

c) in response to detecting that the error ratio for the target drive extent exceeds a threshold error ratio, perform a proactive copy operation on the target drive extent that copies all host data stored on the target drive extent to a newly allocated drive extent, wherein performing the proactive copy operation on the target drive extent also modifies a RAID extent entry in the RAID mapping table that stored an indication of the target drive extent to store an indication of the newly allocated drive extent, whereby the host data copied from the target drive extent to the newly allocated drive extent is accessed by subsequently received I/O operations on the newly allocated drive extent.

11. The data storage system of claim 10, wherein the program code, when executed by the processing circuitry, further causes the processing circuitry to:

perform the proactive copy operation on the target drive extent at least in part by also setting a dead flag corresponding to the target drive extent; and

responsive to the dead flag corresponding to the target drive extent being set, prevent the target drive extent from subsequently being allocated to any RAID mapping table entry in the RAID mapping table.

12. The data storage system of claim 11, wherein the program code, when executed by the processing circuitry, further causes the processing circuitry to:

receive an end of life indication from the data storage drive; and

in response to receiving the end of life indication from the data storage drive, perform proactive copy operations on drive extents located within the data storage drive, wherein the proactive copy operations only copy host data stored on those drive extents located within the data storage drive that are indicated by particular RAID extent entries in the RAID mapping table to newly allocated drive extents, and wherein the proactive copy operations also modify the particular RAID extent entries in the RAID mapping table that stored indications of the drive extents located within the data storage drive to store indications of the newly allocated drive extents, whereby the host data copied from the drive extents located within the data storage drive that were indicated by RAID extent entries in the RAID mapping table to the newly allocated drive extents is subsequently accessed by subsequently received I/O operations on the newly allocated drive extents.

13. The data storage system of claim 12, wherein the program code, when executed by the processing circuitry, further causes the processing circuitry to:

perform proactive copy operations on all drive extents located within the data storage drive at least in part by also setting a dead flag corresponding to each one of the drive extents located within the data storage drive; and

responsive to the dead flag corresponding to each one of the drive extents located within the data storage drive being set, prevent the drive extents located within the data storage drive from subsequently being allocated to any RAID mapping table entry in the RAID mapping table.

14. The data storage system of claim 13, wherein the program code, when executed by the processing circuitry, further causes the processing circuitry to perform the proactive copy operation on all drive extents located within the data storage drive by further causing the processing circuitry to:

set an end of life flag corresponding to each one of the drive extents that are located within the data storage drive and that are also indicated by RAID extent entries in the RAID mapping table;

detect, by a background process, the set end of life flags corresponding to the drive extents that are located within the data storage drive and that are also indicated by RAID extent entries in the RAID mapping table; and

in response to detecting, by the background process, the set end of life flags corresponding to the drive extents that are located within the data storage drive and that are also indicated by RAID extent entries in the RAID mapping table, copy, by the background process, all the host data stored on drive extents that are located within the data storage drive and that are also indicated by RAID extent entries in the RAID mapping table to the newly allocated drive extents, and modify, by the background process, each one of the RAID extent entries in the RAID mapping table that stored an indication of one of the drive extents located within the data storage drive to instead store an indication of one of the newly allocated drive extents.

15. The data storage system of claim 12, wherein the data storage drive within which the target drive extent is located includes a set of reserved disk sectors that are used to replace failed disk sectors;

and

wherein the received completion status for that I/O operation that indicates that a soft media error occurred within the data storage drive while performing that I/O operation on the target drive extent indicates that one of the set of reserved disk sectors was used to replace an original disk sector to which that I/O operation was directed within the target drive extent, in response to a failure of the original disk sector.

16. The data storage system of claim 15, wherein the end of life indication received from the data storage drive indicates that a threshold maximum number of the reserved disk sectors have been used to replace original disk sectors to which I/O operations were directed, in response to failures of the original disk sectors.

17. The data storage system of claim 16, wherein the threshold maximum number of the reserved disk sectors comprises the total number of reserved disk sectors in the data storage drive.

18. The data storage system of claim 11, wherein the program code, when executed by the processing circuitry, further causes the processing circuitry to perform the proactive copy operation on the target drive extent at least in part by further causing the processing circuitry to:

set an end of life flag corresponding to the target drive extent;

detect, by a background process, the set end of life flag corresponding to the target drive extent; and

in response to detecting the set end of life flag corresponding to the target drive extent, copy, by the background process, all the host data stored on the target drive extent to the newly allocated drive extent, and modifying, by the background process, the RAID extent entry in the RAID mapping table that stored an indication of the target drive extent to store an indication of the newly allocated drive extent.

19. A non-transitory computer readable medium for providing RAID (Redundant Array of Independent Disks) data protection for a storage object in a data storage system, wherein the data storage system includes a storage processor and an array of data storage drives communicably coupled to the storage processor, the non-transitory computer readable medium comprising instructions stored thereon that when executed on processing circuitry in the storage processor perform the steps of:

generating a RAID mapping table, wherein the RAID mapping table contains a plurality of RAID extent entries, wherein each RAID extent entry contained in the RAID mapping table indicates a predetermined total number of drive extents that each persistently store host data written to a corresponding one of a plurality of RAID extents within a logical address space that is mapped to the storage object, wherein each drive extent comprises a unique contiguous region of non-volatile data storage located on one of the data storage drives, and wherein each one of the data storage drives has multiple drive extents located thereon;

for each I/O operation directed to the storage object, performing a monitoring operation by the storage processor, wherein the monitoring operation includes;

i) incrementing a total I/O operations counter corresponding to a target drive extent to which that I/O operation is directed, wherein the total I/O operations counter corresponding to the target drive extent stores a total number of I/O operations that have been directed to the target drive extent,

ii) receiving, from a data storage drive within which the target drive extent is located, a completion status for that I/O operation, and

iii) in response to detecting that the received completion status for that I/O operation indicates that a soft media error occurred within the data storage drive while performing that I/O operation on the target drive extent;

a) incrementing a soft media error counter corresponding to the target drive extent, wherein the soft media error counter corresponding to the target drive extent stores a total number of soft media errors that have occurred while performing I/O operations on the target drive extent,

b) calculating an error ratio for the target drive extent, wherein the error ratio for the target drive extent comprises a ratio of a current value of the soft media error counter corresponding to the target drive extent to a current value of the total I/O operations counter corresponding to the target drive extent, and

c) in response to detecting that the error ratio for the target drive extent exceeds a threshold error ratio, performing a proactive copy operation on the target drive extent that copies all host data stored on the target drive extent to a newly allocated drive extent, wherein performing the proactive copy operation on the target drive extent also modifies a RAID extent entry in the RAID mapping table that stored an indication of the target drive extent to store an indication of the newly allocated drive extent, whereby the host data copied from the target drive extent to the newly allocated drive extent is accessed by subsequently received I/O operations on the newly allocated drive extent.