Storage system and folding method for enhanced performance
Rather than use one long folding operation to fold data from single-level cell (SLC) blocks into a multi-level cell (MLC) block, a storage system uses a multi-stage folding operation. By breaking up the folding process into stages, SLC blocks involved in an earlier stage can be released prior to a subsequent stage being performed. This can increase performance of the storage system by releasing SLC source blocks sooner and reducing an SLC block budget requirement.
1. A storage system comprising:
a memory comprising a plurality of single-level cell (SLC) blocks and a multi-level cell (MLC) block; and
a controller coupled to the memory and configured to fold data stored in the plurality of SLC blocks into the MLC block using a multi-stage process;
wherein the controller is further configured to:
perform a first stage of the multi-stage process, wherein in the first stage, a first subset of the plurality of SLC blocks is folded into a first subset of pages of the MLC block; and
in response to successful completion of the first stage of the multi-stage process:
perform a second stage of the multi-stage process, wherein in the second stage, a second subset of the plurality of SLC blocks is folded into a second subset of pages of the MLC block, wherein the second stage is performed without reading data written in the first stage and writing it back in the second stage; and
release the first subset of the plurality of SLC blocks between the first and second stages.
2. The storage system of claim 1 , wherein the controller is further configured to perform the first stage irrespective of validity of data in the first subset of SLC blocks.
3. The storage system of claim 1 , wherein the first and second stages are performed sequentially at different times and not in parallel.
4. The storage system of claim 1 , wherein:
the MLC block comprises a quad-level cell (QLC) block;
in the first stage, two SLC blocks are folded into lower and middle pages of the QLC block; and
in the second stage, two other SLC blocks are folded into upper and top pages of the QLC block.
5. The storage system of claim 1 , wherein:
the MLC block comprises a quad-level cell (QLC) block; and
in the first stage, a first SLC block is written into lower and middle pages of the QLC block, then released before a second SLC block is written into the lower and middle pages of the QLC block.
6. The storage system of claim 5 , wherein:
in the second stage, a third SLC block is written into upper and top pages of the QLC block, then released before a fourth SLC block is written into the upper and top pages of the QLC block.
7. The storage system of claim 1 , wherein:
the MLC block comprises a quad-level cell (QLC) block; and
in the second stage, one SLC block is written into upper and top pages of the QLC block, then released before another SLC block is written into the upper and top pages of the QLC block.
8. The storage system of claim 1 , wherein the controller is further configured to fold the data in response to a number of available SLC blocks in the memory being below a threshold.
9. The storage system of claim 1 , wherein the controller is further configured to verify that the first subset of the plurality of SLC blocks was successfully written in the first subset of pages of the MLC block before releasing the first subset of the plurality of SLC blocks.
10. The storage system of claim 1 , wherein the controller is further configured to release the second subset of the plurality of SLC blocks after the second stage.
11. The storage system of claim 1 , wherein the memory comprises a three-dimensional memory.
12. In a storage system comprising a memory comprising a plurality of single-level cell (SLC) blocks and a quad-level cell (QLC) block, a method comprising:
determining how many SLC blocks are available for folding; and
in response to determining that two SLC blocks are available for folding:
determining whether lower and middle pages of the QLC block have been programmed;
in response to determining that the lower and middle pages of the QLC block have not been programmed;
programming the lower and middle pages of the QLC block with the two SLC blocks; and then
in response to the lower and middle pages of the QLC block being programmed successfully:
releasing the two SLC blocks; and then programming upper and top pages of the QLC block; and
in response to determining that the lower and middle pages of the QLC block have been programmed, programming the upper and top pages of the QLC block with the two SLC blocks.
13. The method of claim 12 , wherein the upper and top pages of the QLC block are programmed without reading data written in the lower and middle pages of the QLC block and using that data in programming the upper and top pages of the QLC block.
14. The method of claim 12 , wherein the lower and middle pages of the QLC block are programmed with the two SLC blocks irrespective of a validity of data stored in the two SLC blocks.
15. The method of claim 12 , further comprising:
in response to determining that four SLC blocks are available for folding, fold the four SLC blocks into the QLC block.
16. The method of claim 12 , wherein determining how many SLC blocks are available for folding is performed in response to determining that a number of available SLC blocks in the memory is below a threshold.
17. The method of claim 12 , wherein the two SLC blocks are released in response to verifying that the lower and middle pages have been successfully written.
18. The storage system of claim 17 , further comprising using an enhanced post-write read (EPWR) operation to verify that the lower and middle pages have been successfully written.
19. The method of claim 12 , further comprising releasing the two SLC blocks after the upper and top pages have been programmed.
20. A storage system comprising:
a memory comprising a plurality of single-level cell (SLC) blocks and a multi-level cell (MLC) block; and
means for folding data stored in the plurality of SLC blocks into the MLC block using a multi-stage process;
wherein:
in a first stage of the multi-stage process, a first subset of the plurality of SLC blocks is folded into a first subset of pages of the MLC block;
in a second stage of the multi-stage process, which is performed in response to successful completion of the first stage, a second subset of the plurality of SLC blocks is folded into a second subset of pages of the MLC block, wherein the second stage is performed without reading data written in the first stage and writing it back in the second stage; and
the first subset of the plurality of SLC blocks is released between the first and second stages.