Non-volatile memory-based mass storage devices and methods for writing data thereto
View Patent ↗A non-volatile solid state memory-based mass storage device having at least one non-volatile memory component and methods of operating the storage device. In one aspect of the invention, the one or more memory components define a memory space partitioned into user memory and over-provisioning pools based on a P/E cycle count stored in a block information record. The storage device transfers the P/E cycle count of erased blocks to a host and the host stores the P/E cycle count in a content addressable memory. During a host write to the storage device, the host issues a low P/E cycle count number as a primary address to the content addressable memory, which returns available block addresses of blocks within the over-provisioning pool as a first dimension in a multidimensional address space. Changed files are preferably updated in append mode and the previous version can be maintained for version control.
1. A NAND flash-based mass storage device, the storage device comprising:
a substrate;
at least two NAND flash memory components on the substrate, the memory components defining a NAND flash memory space organized into blocks and pages, each block having a block information record storing the number of program and erase cycles of the block;
a NAND flash controller including a flash translation layer, the flash translation layer translating logical block addresses into physical block addresses; and
an interface that interfaces the storage device with a host system through a system interface;
wherein the storage device sends the block information record of at least some blocks to the host, the host stores the block information record of the program and erase cycles in a content addressable memory, the content addressable memory associates the logical address of the block with the number of program and erase cycles and the flash translation layer translating the logical address to a physical block address, and, on writing data to the storage device, the host selects a number of program and erase cycles to obtain a logical block address from the content addressable memory the file translation layer translating the logical address into a physical address of a block to which the data are written.
2. The storage device of claim 1 , wherein the content addressable memory is a ternary content addressable memory.
3. The storage device of claim 1 , wherein the content addressable memory is a look-up table.
4. The storage device of claim 1 , wherein the content addressable memory is a translation lookaside buffer.
5. The storage device of claim 1 , wherein the interface conforms to the SATA standard.
6. The storage device of claim 1 , wherein the interface is a multi-lane PCIe interface.
7. The storage device of claim 1 , wherein the interface comprises at least two independent links connected to a multi-ported flash translation layer that acts as crossbar switch for addressing blocks in a unified NAND flash memory pool.
8. The storage device of claim 1 , wherein the interface comprises at least two independent links and wherein each link connects to its own NAND flash memory domain and link-selection by individual threads is done on the level of the host.
9. A method of operating the storage device of claim 1 , the method comprising monitoring the program and erase count of all blocks, partitioning the NAND flash memory space into a user memory pool and an over-provisioning pool, and assigning one or more blocks with the lowest program and erase count after an erase cycle to the over-provisioning pool.
10. The method of claim 9 , wherein blocks assigned to the over-provisioning pool are written to in single level cell mode, the majority of blocks in the user memory pool are configured in multilevel cell mode, and space reclamation uses multi-level cell mode blocks for coalescence of valid data.
11. The method of claim 9 , wherein blocks assigned to the over-provisioning pool are formatted to one of a plurality of mapping unit sizes.
12. The method of claim 9 , wherein some files are stored in a mapping unit and treated as a single object by the file system, and other files are stored in a plurality of mapping units and treated as composite objects.
13. The method of claim 9 , wherein files changed by the host are stored as partial updates changed parts of the files are stored in a new physical location, and metadata are updated to point to the new physical location of the partial update.
14. The method of claim 9 wherein, after an original file is updated, the original file is not invalidated but assigned a version control level.