Mapped raid (redundant array of independent disks) with drive health aware protection groups
Mapped RAID protection is provided for data written into multiple logical RAID spaces, and the data written to each one of the RAID spaces is stored in data storage drives contained in a corresponding one of multiple protection groups of data storage drives. Drive health information is collected for each data storage drive contained in the protection groups, and a faultiness level of each individual data storage drive in the protection groups is calculated based on the drive health information. A faultiness-balanced distribution of the data storage drives across the protection groups is then generated based on the calculated faultiness levels of the individual data storage drives.
1. A method comprising:
providing mapped RAID (Redundant Array of Independent Disks) protection for data written into a plurality of logical RAID spaces, wherein data written to each one of the logical RAID spaces is stored in data storage drives contained in a corresponding one of a plurality of protection groups of data storage drives;
collecting drive health information for each data storage drive contained in the protection groups;
calculating a faultiness level of each individual data storage drive in the protection groups based on the drive health information; and
generating a faultiness-balanced distribution of the data storage drives across the protection groups based on the calculated faultiness levels of the individual data storage drives, at least in part by determining a faultiness-balanced distribution of the data storage drives across the protection groups that minimizes a range of values of a faultiness metric calculated for the individual protection groups based on the faultiness levels of the individual data storage drives, wherein the values of the faultiness metric calculated for the individual protection groups is based on the sizes of the individual protection groups, and wherein the faultiness metric calculated for each protection group comprises a size of the protection group divided by a sum of the faultiness levels of the data storage drives contained in the protection group.
2. The method of claim 1 , further comprising:
determining whether the faultiness-balanced distribution of the data storage drives across the protection groups matches a current distribution of the data storage drives across the protection groups; and
in response to the faultiness-balanced distribution of the data storage drives across the protection groups not matching the current distribution of the data storage drives across the protection groups, automatically redistributing the data storage drives across the protection groups according to the faultiness-balanced distribution of the data storage drives.
3. The method of claim 2 , wherein automatically redistributing the data storage drives across the protection groups of data storage drives according to the faultiness-balanced distribution of the data storage drives includes moving at least one of data storage drive from a first one of the protection groups to a second one of the protection groups.
4. The method of claim 1 , wherein the sizes of the individual protection groups used to calculate the values of the faultiness metric for the individual protection groups comprises a total capacity of each protection group.
5. The method of claim 1 , wherein the sizes of the individual protection groups used to calculate the values of the faultiness metric for the individual protection groups comprises an amount of data stored in each protection group.
6. The method of claim 1 , further comprising:
determining whether the faultiness-balanced distribution of the data storage drives across the protection groups provides a significant improvement over a current distribution of the data storage drives across the protection groups; and
in response to the faultiness-balanced distribution of the data storage drives across the protection groups providing a significant improvement over the current distribution of the data storage drives across the protection groups, automatically redistributing the data storage drives across the protection groups according to the faultiness-balanced distribution of the data storage drives.
7. A data storage system comprising:
at least one storage processor including processing circuitry and a memory;
a plurality of data storage drives communicably coupled to the 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:
provide mapped RAID (Redundant Array of Independent Disks) protection for data written into a plurality of logical RAID spaces, wherein data written to each one of the logical RAID spaces is stored in data storage drives contained in a corresponding one of a plurality of protection groups of data storage drives,
collect drive health information for each data storage drive contained in the protection groups,
calculate a faultiness level of each individual data storage drive in the protection groups based on the drive health information, and
generate a faultiness-balanced distribution of the data storage drives across the protection groups based on the calculated faultiness levels of the individual data storage drives, at least in part by causing the processing circuitry to determine a faultiness-balanced distribution of the data storage drives across the protection groups that minimizes a range of values of a faultiness metric calculated for the individual protection groups based on the faultiness levels of the individual data storage drives, wherein the values of the faultiness metric calculated for the individual protection groups is based on the sizes of the individual protection groups, and wherein the faultiness metric calculated for each protection group comprises a size of the protection group divided by a sum of the faultiness levels of the data storage drives contained in the protection group.
8. The data storage system of claim 7 , wherein the program code, when executed by the processing circuitry, further causes the processing circuitry to:
determine whether the faultiness-balanced distribution of the data storage drives across the protection groups matches a current distribution of the data storage drives across the protection groups; and
in response to the faultiness-balanced distribution of the data storage drives across the protection groups not matching the current distribution of the data storage drives across the protection groups, automatically redistribute the data storage drives across the protection groups according to the faultiness-balanced distribution of the data storage drives.
9. The data storage system of claim 8 , wherein the program code, when executed by the processing circuitry, further causes the processing circuitry to automatically redistribute the data storage drives across the protection groups of data storage drives according to the faultiness-balanced distribution of the faulty data storage drives at least in part by causing the processing circuitry to move at least one of data storage drive from a first one of the protection groups to a second one of the protection groups.
10. The data storage system of claim 7 , wherein the sizes of the individual protection groups used to calculate the values of the faultiness metric for the individual protection groups comprises a total capacity of each protection group.
11. The data storage system of claim 7 , wherein the sizes of the individual protection groups used to calculate the values of the faultiness metric for the individual protection groups comprises an amount of data stored in each protection group.
12. The data storage system of claim 7 , wherein the program code, when executed by the processing circuitry, further causes the processing circuitry to:
determine whether the faultiness-balanced distribution of the data storage drives across the protection groups provides a significant improvement over a current distribution of the data storage drives across the protection groups; and
in response to the faultiness-balanced distribution of the data storage drives across the protection groups providing a significant improvement over the current distribution of the data storage drives across the protection groups, automatically redistribute the data storage drives across the protection groups according to the faultiness-balanced distribution of the data storage drives.
13. A computer program product including a non-transitory computer readable medium having instructions stored thereon, wherein the instructions, when executed on processing circuitry, cause the processing circuitry to perform the steps of:
providing mapped RAID (Redundant Array of Independent Disks) protection for data written into a plurality of logical RAID spaces, wherein data written to each one of the logical RAID spaces is stored in data storage drives contained in a corresponding one of a plurality of protection groups of data storage drives;
collecting drive health information for each data storage drive contained in the protection groups;
calculating a faultiness level of each individual data storage drive in the protection groups based on the drive health information; and
generating a faultiness-balanced distribution of the data storage drives across the protection groups based on the calculated faultiness levels of the individual data storage drives, at least in part by determining a faultiness-balanced distribution of the data storage drives across the protection groups that minimizes a range of values of a faultiness metric calculated for the individual protection groups based on the faultiness levels of the individual data storage drives, wherein the values of the faultiness metric calculated for the individual protection groups is based on the sizes of the individual protection groups, and wherein the faultiness metric calculated for each protection group comprises a size of the protection group divided by a sum of the faultiness levels of the data storage drives contained in the protection group.