Multi-BCRC raid protection for CKD
A storage array that presents a logical storage device to hosts that is accessed using front-end tracks and access tangible managed drive using back-end tracks locates multiple front-end tracks in individual back-end tracks. Error-correcting codes are used to identify different front-end tracks in a back-end track when the back-end track is copied from the managed drives into storage array memory. CKD front-end tracks can be split into multiple partial CKD front-end tracks that are located at contiguous address space in different back-end tracks. The front-end tracks that are located in a particular back-end track may be selected to reduce or minimize unused space. The front-end tracks in a back-end track may be logically stored on different production volumes.
1. An apparatus comprising:
a data storage system comprising:
a plurality of non-volatile drives; and
a plurality of interconnected compute nodes that present at least one logical production volume to hosts and manage access to the plurality of non-volatile drives, wherein the hosts access the production volume using front-end tracks as allocation units and the compute nodes access the plurality of non-volatile drives using back-end tracks as allocation units and the compute nodes locate multiple front-end tracks in a single back-end track;
whereby storage capacity of the plurality of non-volatile drives is more efficiently utilized in comparison with locating only one front-end track in each back-end track.
2. The apparatus of claim 1 wherein the front-end tracks are count key data (CKD) tracks.
3. The apparatus of claim 2 wherein the compute nodes split at least some of the CKD front-end track into multiple partial CKD front-end tracks that are located in separate back-end tracks.
4. The apparatus of claim 3 wherein the compute nodes locate one of the partial CKD front-end tracks and multiple CKD front-end tracks in one back-end track.
5. The apparatus of claim 3 wherein the compute nodes locate one of the partial CKD front-end tracks and multiple fixed-block architecture (FBA) front-end tracks in one back-end track.
6. The apparatus of claim 1 wherein the compute nodes locate front-end tracks from different production volumes in a back-end track.
7. The apparatus of claim 1 wherein the compute nodes use error-correcting codes to identify different front-end tracks in a back-end track.
8. A method comprising:
in a data storage system comprising a plurality of non-volatile drives and a plurality of interconnected compute nodes:
the compute nodes presenting at least one logical production volume to hosts and managing access to the plurality of non-volatile drives, wherein the hosts access the production volume using front-end tracks as allocation units and the compute nodes access the plurality of non-volatile drives using back-end tracks as allocation units; and
the compute nodes locating multiple front-end tracks in a single back-end track;
whereby storage capacity of the plurality of non-volatile drives is more efficiently utilized in comparison with locating only one front-end track in each back-end track.
9. The method of claim 8 comprising the compute nodes locating multiple count key data (CKD) front-end tracks in the single back-end track.
10. The method of claim 9 comprising the compute nodes splitting at least some of the CKD front-end track into multiple partial CKD front-end tracks and locating those partial CKD front-end tracks in separate back-end tracks.
11. The method of claim 10 comprising the compute nodes locating one of the partial CKD front-end tracks and multiple CKD front-end tracks in one back-end track.
12. The method of claim 10 comprising the compute nodes locating one of the partial CKD front-end tracks and multiple fixed-block architecture (FBA) front-end tracks in one back-end track.
13. The method of claim 8 comprising the compute nodes locating front-end tracks from different production volumes in a back-end track.
14. The method of claim 8 comprising the compute nodes using error-correcting codes to identify different front-end tracks in a back-end track.
15. A computer-readable storage medium storing instructions that when executed by a computer cause the computer to perform a method for using a computer system to store data, the method comprising:
compute nodes presenting at least one logical production volume to hosts, wherein the hosts access the production volume using front-end tracks as allocation units and the compute nodes access non-volatile drives using back-end tracks as allocation units; and
the compute nodes locating multiple front-end tracks in a single back-end track; whereby storage capacity of the non-volatile drives is more efficiently utilized in comparison with locating only one front-end track in each back-end track.
16. The computer-readable storage medium of claim 15 wherein the method further comprises the compute nodes using error-correcting codes to identify different front-end tracks in a back-end track.
17. The computer-readable storage medium of claim 15 wherein the method further comprises the compute nodes locating multiple count key data (CKD) front-end tracks in the single back-end track.
18. The computer-readable storage medium of claim 17 wherein the method further comprises the compute nodes splitting at least some of the CKD front-end track into multiple partial CKD front-end tracks and locating those partial CKD front-end tracks in separate back-end tracks.
19. The computer-readable storage medium of claim 18 wherein the method further comprises the compute nodes locating one of the partial CKD front-end tracks and multiple CKD front-end tracks in one back-end track.
20. The computer-readable storage medium of claim 18 wherein the method further comprises the compute nodes locating one of the partial CKD front-end tracks and multiple fixed-block architecture (FBA) front-end tracks in one back-end track.