Storage array disk recovery
One or more aspects of the present disclosure relate to recovering at least one failed disk. In embodiments, determining a storage reserve capacity allocated for recovering at least one storage device of a storage array is determined. Zero or more storage portions from each storage device of at least one storage cluster for disk recovery are adaptively assigned based on the storage reserve capacity. The failing and/or failed disk using the assigned storage portions is recovered in response to detecting a failing and/or failed disk.
1. A method comprising:
determining a storage reserve capacity allocated for recovering at least one storage device of a plurality of storage devices in a storage cluster corresponding to a storage array;
based on the storage reserve capacity, adaptively assigning zero or more storage portions from each storage device in a subset of the plurality of storage devices in the storage cluster for disk recovery, wherein assigning the portions includes:
assigning at most one portion from each storage device in the subset of the plurality of storage devices as a spare for disk recovery, wherein the at most one portion is assigned based on a spare distribution pattern corresponding to a logical matrix representation of the storage cluster,
allowing at least one storage device of the plurality of storage devices to remain free from having any portions assigned as disk recovery spares, and
in response to expanding the storage cluster to include another storage cluster, redistributing spares for disk recovery such that the storage cluster and the other storage cluster include an equal number of spare partitions; and
in response to detecting a failing and/or failed disk, recovering the failing and/or failed disk using the assigned storage portions.
2. The method of claim 1 further comprising:
adaptively assigning the zero or more storage portions by using a successive redistribution and assignment technique.
3. The method of claim 1 further comprising:
subdividing each storage device into a plurality of partitions, wherein the partitions are equal-sized;
establishing at least one redundant array of independent disks (RAID) with a width (W) corresponding to RAID data (D) and parity (P) members; and
establishing the partitions amount based on a multiple of the RAID width.
4. The method of claim 3 further comprising:
establishing the at least one storage cluster based on the RAID width and the partitions amount.
5. The method of claim 4 further comprising:
establishing at least one storage matrix defined by a number of storage devices and the partitions amount, wherein the number of storage devices corresponds to W+1, wherein the storage cluster corresponds to the at least one storage matrix.
6. The method of claim 5 further comprising:
establishing a set of storage submatrices within the storage cluster, wherein each sub-matrix is defined by the number of storage devices and a partition sub-matrix amount corresponding to the RAID width.
7. The method of claim 6 further comprising:
provisioning at least one RAID group from zero or more partitions defined by each sub-matrix, wherein the at least one RAID group's storage capacity corresponds to the storage reserve capacity.
8. The method of claim 7 further comprising:
provisioning each RAID group using an adaptive successive redistribution and assignment technique (adaptive technique), wherein the adaptive technique includes selecting at most one storage device's partition in each storage sub-matrix; and
establishing an amount of RAID groups based on a maximum RAID group capacity and the allocated storage reserve capacity.
9. The method of claim 8 further comprising:
determining a storage cluster amount; and
based on the storage cluster amount, provisioning each cluster with an equal amount of RAID groups.
10. The method of claim 9 further comprising:
exclusively configuring each RAID group's partitions for disk recovery.
11. An apparatus configured to a memory and at least one processor configured to:
determine a storage reserve capacity allocated for recovering at least one storage device of a plurality of storage devices in a storage cluster corresponding to a storage array;
based on the storage reserve capacity, adaptively assign zero or more storage portions from each storage device in a subset of the plurality of storage devices in the storage cluster for disk recovery, wherein assigning the portions includes:
assigning at most one portion from each storage device in the subset of the plurality of storage devices as a spare for disk recovery, wherein the at most one portion is assigned based on a spare distribution pattern corresponding to a logical matrix representation of the storage cluster,
allowing at least one storage device of the plurality of storage devices to remain free from having any portions assigned as disk recovery spares, and
in response to expanding the storage cluster to include another storage cluster, redistributing spares for disk recovery such that the storage cluster and the other storage cluster include an equal number of spare partitions; and
in response to detecting a failing and/or failed disk, recover the failing and/or failed disk using the assigned storage portions.
12. The apparatus of claim 11 further configured to:
adaptively assign the zero or more storage portions by using a successive redistribution and assignment technique.
13. The apparatus of claim 11 further configured to:
subdivide each storage device into a plurality of partitions, wherein the partitions are equal-sized;
establish at least one redundant array of independent disks (RAID) with a width (W) corresponding to RAID data (D) and parity (P) members; and
establish a partitions amount based on a multiple of the RAID width.
14. The apparatus of claim 13 further configured to:
establish the at least one storage cluster based on the RAID width and the partitions amount.
15. The apparatus of claim 14 further configured to:
establish at least one storage matrix defined by a number of storage devices and the partitions amount, wherein the number of storage devices corresponds to W+1, wherein the storage cluster corresponds to the at least one storage matrix.
16. The apparatus of claim 15 further configured to:
establish a set of storage submatrices within the storage cluster, wherein each sub-matrix is defined by the number of storage devices and a partition sub-matrix amount corresponding to the RAID width.
17. The apparatus of claim 16 further configured to:
provision at least one RAID group from zero or more partitions defined by each sub-matrix, wherein the at least one RAID group's storage capacity corresponds to the storage reserve capacity.
18. The apparatus of claim 17 further configured to:
provision each RAID group using an adaptive successive redistribution and assignment technique (adaptive technique), wherein the adaptive technique includes selecting at most one storage device's partition in each storage sub-matrix; and
establish an amount of RAID groups based on a maximum RAID group capacity and the allocated storage reserve capacity.
19. The apparatus of claim 18 further configured to:
determine a storage cluster amount; and
based on the storage cluster amount, provision each cluster with an equal amount of RAID groups.
20. The apparatus of claim 19 further configured to:
exclusively configure each RAID group's partitions for disk recovery.