System and method of data writes and mapping of data for multiple sub-drives
A system and method is disclosed for managing data in a non-volatile memory. The system may include a non-volatile memory having multiple non-volatile memory sub-drives. A controller of the memory system is configured to route incoming host data to a desired sub-drive, keep data within the same sub-drive as its source during a garbage collection operation, and re-map data between sub-drives, separate from any garbage collection operation, when a sub-drive overflows its designated amount logical address space. The method may include initial data sorting of host writes into sub-drives based on any number of hot/cold sorting functions. In one implementation, the initial host write data sorting may be based on a host list of recently written blocks for each sub-drive and a second write to a logical address encompassed by the list may trigger routing the host write to a hotter sub-drive than the current sub-drive.
1. A data storage system, comprising:
a plurality of sub-drives; and
one or more controllers configured to cause:
writing data into a first sub-drive of the plurality of sub-drives;
when a garbage collection operation is triggered:
selecting one sub-drive of the plurality of sub-drives, based on an amount of over-provisioning;
selecting a superblock within the selected one sub-drive, based on an amount of valid data;
copying the valid data in the selected superblock of the selected one sub-drive, to another superblock within only the same selected one sub-drive, without crossing sub-drive boundaries, wherein the another superblock is different from the selected superblock; and
adding an entry, associated with the selected superblock, to a free block list; and
performing re-mapping of a given superblock within a given sub-drive of the plurality of sub-drives to another sub-drive, wherein the another sub-drive is different from the given sub-drive.
2. The data storage system of claim 1 ,
wherein the performing re-mapping comprises:
determining that the given sub-drive has overflowed its predetermined amount of logical address space; and
based on the determining, re-mapping the given superblock within the given sub-drive to a superblock within the another sub-drive,
wherein the performing re-mapping is carried out without performing garbage collection, and
wherein the free block list indicates currently unused physical superblocks available for use.
3. The data storage system of claim 1 , wherein the selecting one sub-drive comprises selecting the one sub-drive based on a current logical capacity occupied by valid data in the one sub-drive and a current write traffic in the one sub-drive.
4. The data storage system of claim 1 ,
wherein the writing data comprises writing the data into the first sub-drive of the plurality of sub-drives when an address of the data is represented in a list associated with a second sub-drive of the plurality of sub-drives, and
wherein the first sub-drive is associated with a data temperature range hotter than the second sub-drive.
5. The data storage system of claim 4 , wherein the list is associated with recent data writes for the second sub-drive.
6. The data storage system of claim 4 , wherein the writing data comprises writing the data into the second sub-drive when the address of the data is not represented in the list.
7. The data storage system of claim 4 , wherein the list comprises most recent host data writes and most recent garbage collection writes.
8. The data storage system of claim 4 , wherein the one or more controllers are configured to cause:
removing an oldest entry off an end of the list and adding a new entry at a beginning of the list when a logical address of the data is absent from the list and when the list is filled to a fixed length of entries.
9. A machine-implemented method, comprising:
writing data into a first sub-drive of a plurality of sub-drives;
when a garbage collection operation is triggered:
selecting one sub-drive of the plurality of sub-drives, based on an amount of over-provisioning;
selecting a superblock within the selected one sub-drive, based on an amount of valid data;
copying the valid data in the selected superblock of the selected one sub-drive, to another superblock within only the same selected one sub-drive, without crossing sub-drive boundaries, wherein the another superblock is different from the selected superblock; and
adding an entry, associated with the selected superblock, to a free block list; and
performing re-mapping of a given superblock within a given sub-drive of the plurality of sub-drives to another sub-drive, wherein the another sub-drive is different from the given sub-drive.
10. The machine-implemented method of claim 9 ,
wherein the performing re-mapping comprises:
determining that the given sub-drive has overflowed its predetermined amount of logical address space; and
based on the determining, re-mapping the given superblock within the given sub-drive to a superblock within the another sub-drive, and
wherein the performing re-mapping is carried out without performing garbage collection.
11. The machine-implemented method of claim 9 , wherein the selecting one sub-drive comprises selecting the one sub-drive based on a current logical capacity occupied by valid data in the one sub-drive and a current write traffic in the one sub-drive.
12. The machine-implemented method of claim 9 ,
wherein the writing data comprises writing the data into the first sub-drive of the plurality of sub-drives when an address of the data is represented in a list associated with a second sub-drive of the plurality of sub-drives, and
wherein the first sub-drive is associated with a data temperature range hotter than the second sub-drive.
13. The machine-implemented method of claim 12 , wherein the list is associated with recent data writes for the second sub-drive.
14. The machine-implemented method of claim 12 , wherein the writing data comprises writing the data into the second sub-drive when the address of the data is not represented in the list.
15. The machine-implemented method of claim 12 , wherein the list comprises most recent host data writes and most recent garbage collection writes.
16. An apparatus, comprising:
means for writing data into a first sub-drive of a plurality of sub-drives;
when a garbage collection operation is triggered:
means for selecting one sub-drive of the plurality of sub-drives, based on an amount of over-provisioning;
means for selecting a superblock within the selected one sub-drive, based on an amount of valid data;
means for copying the valid data in the selected superblock of the selected one sub-drive, to another superblock within only the same selected one sub-drive, without crossing sub-drive boundaries, wherein the another superblock is different from the selected superblock; and
means for adding an entry, associated with the selected superblock, to a free block list; and
means for performing re-mapping of a given superblock within a given sub-drive of the plurality of sub-drives to another sub-drive, wherein the another sub-drive is different from the given sub-drive.
17. The apparatus of claim 16 ,
wherein the means for performing re-mapping comprises:
means for determining that the given sub-drive has overflowed its predetermined amount of logical address space; and
means for re-mapping, based on the determining, the given superblock within the given sub-drive to a superblock within the another sub-drive, and
wherein the means for performing re-mapping is to be carried out without performing garbage collection.
18. The apparatus of claim 16 , wherein the means for selecting one sub-drive comprises means for selecting the one sub-drive based on a current logical capacity occupied by valid data in the one sub-drive and a current write traffic in the one sub-drive.
19. The apparatus of claim 16 ,
wherein the means for writing data comprises means for writing the data into the first sub-drive of the plurality of sub-drives when an address of the data is represented in a list associated with a second sub-drive of the plurality of sub-drives, and
wherein the first sub-drive is associated with a data temperature range hotter than the second sub-drive.
20. The apparatus of claim 19 , wherein the list is associated with recent data writes for the second sub-drive.