Group-based RAID-1 implementation in multi-RAID configured storage array
A storage system that supports multiple RAID levels presents storage objects with front-end tracks corresponding to back-end tracks on non-volatile drives and accesses the drives using a single type of back-end allocation unit that is larger than a back-end track. When the number of members of a protection group of a RAID level does not align with the back-end allocation unit, multiple back-end tracks are grouped and accessed using a single IO. The number of back-end tracks in a group is selected to align with the back-end allocation unit size. If the front-end tracks are variable size, then front-end tracks may be destaged into a smaller number of grouped back-end tracks in a single IO.
1. A method implemented in a data storage system with a plurality of non-volatile drives and a plurality of interconnected compute nodes that access the drives using only one size of back-end allocation unit, the method comprising:
the compute nodes presenting first and second logical production volumes to hosts, wherein front-end tracks of the first and second production volumes correspond to back-end tracks of the non-volatile drives, and the backend allocation unit has a size that is an integer multiple of the back-end track size that is greater than 1;
implementing a first RAID level for the first production volume, wherein protection groups of the first RAID level have a number of back-end track members that aligns with the backend allocation unit; and
implementing a second RAID level for the second production volume, wherein protection groups of the second RAID level have a number of back-end track members that does not align with the backend allocation unit, the compute nodes configured to respond to hosts accessing front-end tracks of the second production volume by grouping a plurality of accessed back-end tracks and accessing the back-end tracks of the group using a single IO.
2. The method of claim 1 wherein implementing the first RAID level for the first production volume comprises implementing RAID-1 for the first production volume.
3. The method of claim 2 wherein grouping a plurality of accessed back-end tracks comprises grouping four accessed back-end tracks.
4. The method of claim 3 wherein accessing the back-end tracks of the group using a single IO comprises copying the grouped back-end tracks from the non-volatile drives to cache.
5. The method of claim 3 wherein accessing the back-end tracks of the group using a single IO comprises destaging the grouped back-end tracks from cache to the non-volatile drives.
6. The method of claim 3 comprising destaging a larger number of variable size front-end tracks into a smaller number of grouped back-end tracks.
7. The method of claim 3 comprising destaging a number of fixed size front-end tracks into an equal number of grouped back-end tracks.
8. A non-transitory computer-readable storage medium storing instructions that when executed by a compute node of a storage array cause the compute node to perform a method for accessing non-volatile drives using only one size of back-end allocation unit while supporting multiple RAID levels, the method comprising:
presenting first and second logical production volumes to hosts, wherein front-end tracks of the first and second production volumes correspond to back-end tracks of the non-volatile drives, and the backend allocation unit has a size that is an integer multiple of the back-end track size that is greater than 1;
implementing a first RAID level for the first production volume, wherein protection groups of the first RAID level have a number of back-end track members that aligns with the backend allocation unit; and
implementing a second RAID level for the second production volume, wherein protection groups of the second RAID level have a number of back-end track members that does not align with the backend allocation unit, the compute nodes configured to respond to hosts accessing front-end tracks of the second production volume by grouping a plurality of accessed back-end tracks and accessing the back-end tracks of the group using a single IO.
9. The non-transitory computer-readable storage medium of claim 8 wherein the method further comprises implementing the first RAID level for the first production volume comprises implementing RAID-1 for the first production volume.
10. The non-transitory computer-readable storage medium of claim 9 wherein the method further comprises grouping a plurality of accessed back-end tracks comprises grouping four accessed back-end tracks.
11. The non-transitory computer-readable storage medium of claim 10 wherein the method further comprises copying the grouped back-end tracks from the non-volatile drives to cache.
12. The non-transitory computer-readable storage medium of claim 10 wherein the method further comprises destaging the grouped back-end tracks from cache to the non-volatile drives.
13. The non-transitory computer-readable storage medium of claim 10 wherein the method further comprises destaging a larger number of variable size front-end tracks into a smaller number of grouped back-end tracks.
14. The non-transitory computer-readable storage medium of claim 10 wherein the method further comprises destaging a number of fixed size front-end tracks into an equal number of grouped back-end tracks.
15. An apparatus comprising:
a plurality of non-volatile drives;
a plurality of interconnected compute nodes that access the drives using only one size of back-end allocation unit, the compute nodes configured to present first and second logical production volumes to hosts, wherein front-end tracks of the first and second production volumes correspond to back-end tracks of the non-volatile drives, and the backend allocation unit has a size that is an integer multiple of the back-end track size that is greater than 1, and wherein a first RAID level is implemented for the first production volume, wherein protection groups of the first RAID level have a number of back-end track members that aligns with the backend allocation unit, and a second RAID level is implemented for the second production volume, wherein protection groups of the second RAID level have a number of back-end track members that does not align with the backend allocation unit, the compute nodes configured, responsive to hosts accessing front-end tracks of the second production volume, to group a plurality of accessed back-end tracks and access the back-end tracks of the group using a single IO.
16. The apparatus of claim 15 wherein the first RAID level is RAID-1.
17. The apparatus of claim 16 wherein each group contains four accessed back-end tracks.
18. The apparatus of claim 17 wherein the compute nodes copy grouped back-end tracks from the non-volatile drives to cache with a single IO.
19. The apparatus of claim 17 wherein the compute nodes destage grouped back-end tracks from cache to the non-volatile drives with a single IO.
20. The apparatus of claim 17 wherein the compute nodes destage a larger number of variable size front-end tracks into a smaller number of grouped back-end tracks.