# US 2014/0156921 A1 · Methods for writing data to non-volatile memory-based mass storage devices

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2014/0156921 A1
- Document ID: US20140156921A1
- Status: Published application
- Kind: A1
- Application number: 14/175,553
- Filed: 2014-02-07
- Published: 2014-06-05
- Earliest claimed: 2011-10-03
- Assignees: OCZ Storage Solutions Inc.
- Applicants: OCZ Storage Solutions Inc.
- Inventors: Franz Michael Schuette; William Ward Clawson
- Web page: https://patentgrep.com/patent/US20140156921A1

## CPC

- G06F 3/0679: {Non-volatile semiconductor memory device, e.g. flash memory, one time programmable memory [OTP]}
- G06F 3/0616: {in relation to life time, e.g. increasing Mean Time Between Failures [MTBF]}
- G06F 3/0619: {in relation to data integrity, e.g. data losses, bit errors}
- G06F 3/0629: {Configuration or reconfiguration of storage systems}
- G06F 3/064: {Management of blocks}
- G06F 3/0688: {Non-volatile semiconductor memory arrays}
- G11C 16/349: {Arrangements for evaluating degradation, retention or wearout, e.g. by counting erase cycles}
- G06F 12/0246: {in block erasable memory, e.g. flash memory}
- G06F 2212/7201: Logical to physical mapping or translation of blocks or pages
- G06F 2212/7202: Allocation control and policies
- G06F 2212/7204: Capacity control, e.g. partitioning, end-of-life degradation
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G11C: STATIC STORES (semiconductor memory devices H10B)

## Abstract

Methods of operating a non-volatile solid state memory-based mass storage device having at least one non-volatile memory component. 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.

## Claims

### Claim 1

In a computer system comprising a host, a method of storing data in a non-volatile solid state memory-based mass storage device, the mass storage device having a substrate and at least one non-volatile memory component mounted thereon, memory locations in the non-volatile memory component having first order addresses and second order addresses, the method comprising:
generating with the non-volatile memory component a record of the number of programming cycles relating to the first order address;
transmitting the record to the host;
selecting with the host a low number of programming cycles within a range of numbers of programming cycles of records relating to the first order addresses as an initial address for a write access to the non-volatile memory device.

### Claim 2 (depends on 1)

The method of claim 1, wherein the non-volatile memory components are NAND flash memory components and the first order addresses correspond to block addresses.

### Claim 3 (depends on 1)

The method of claim 1, wherein the host issues a low number of the range of numbers of programming cycles to a content addressable memory and the content addressable memory returns the addresses of available blocks.

### Claim 4 (depends on 3)

The method of claim 3, wherein the memory locations corresponding to the first order addresses in the non-volatile memory device comprise a user memory space and an over-provisioning memory space.

### Claim 5 (depends on 4)

The method of claim 4, wherein the number of programming cycles of the first order addresses only from blocks assigned to the over-provisioning pool is transmitted to the host.

### Claim 6 (depends on 5)

The method of claim 5 wherein, after a block is completely written, it is moved into the user memory space.

### Claim 7

A method for writing data from a host computer system to a NAND flash memory-based mass storage device comprising an array of NAND flash memory components, the mass storage device having a substrate and having mounted thereon a connector to interface with the host computer system, the NAND flash memory components, and a controller IC to provide functional connectivity between the NAND flash memory components and the host computer system, each of the NAND flash memory components having blocks and pages and each block having a block information record, the method comprising:
partitioning the array of the NAND flash memory components into a user memory pool and a over-provisioning pool, wherein blocks are dynamically assigned to either pool;
transferring the block information record of the program/erase cycle counts of all blocks in the over-provisioning pool to the host;
storing the block information records in a content addressable memory on the host;
sorting the program/erase cycle counts in the content-addressable memory;
selecting a low number of program/erase cycle count to generate the address of a block in the over-provisioning pool;
writing the data to pages in the selected block;
moving the block from the over-provisioning pool to the user memory pool;
replenishing the over-provisioning pool with an erased block from the user memory pool; and
updating the block information records stored in the content-addressable memory.

### Claim 8 (depends on 7)

The method of claim 7, wherein all host writes are done in append mode and wherein a file update will be stored in a new location and the metadata are updated to specify the new location of the file.

### Claim 9 (depends on 7)

The method of claim 7, further comprising formatting on-demand a block of the NAND flash memory components to have a variable mapping unit size.

### Claim 10 (depends on 7)

The method of claim 7, further comprising treating a file corresponding in size to a mapping unit as an object and treating a file that is larger than a mapping unit as a composite object.

### Claim 11 (depends on 8)

The method of claim 8, wherein a file update only updates a partial file containing the changed portion of the file.

### Claim 12 (depends on 8)

The method of claim 8 wherein, after a file update the previous and the final version are valid data and wherein a version control index specifies the last version.

### Claim 13 (depends on 12)

The method of claim 12, wherein a time-out of a non-final version of a file un-protects the non-final data for space reclamation by the storage device.

### Claim 14

A method for writing data from a host computer system to a NAND flash memory-based mass storage device comprising an array of multi-level cell NAND flash memory components, the mass storage device having a substrate and having mounted thereon a connector to interface with the host computer system, the NAND flash memory components, and a controller IC to provide functional connectivity between the NAND flash memory components and the host computer system, each of the NAND flash memory components having blocks and pages and each block having a block information record, the method comprising:
partitioning the array of NAND flash memory components into a user memory pool and a over-provisioning pool, wherein blocks are dynamically assigned to either pool;
specifying first blocks of the blocks in the over-provisioning pool to only store a single bit in each cell when a write from the host occurs;
transferring the block information record of the program/erase cycle counts of all blocks in the over-provisioning pool to the host;
storing the block information records in a content addressable memory on the host;
selecting a low number of program/erase cycle count to generate the address of a block in the over-provisioning pool;
writing the data to pages in the selected block;
moving the block from the over-provisioning pool to the user memory pool;
replenishing the over-provisioning pool with an erased block from the user memory pool; and
updating the block information records stored in the content-addressable memory.

### Claim 15 (depends on 13)

The method of claim 13, wherein blocks written to by the host and moved to the user memory, can be configured to store two bits per cell of the block.

### Claim 16 (depends on 13)

The method of claim 13 having second blocks in the user memory that are configured to store two bits per cell, those blocks being excluded from being assigned to the over-provisioning pool.

### Claim 17 (depends on 15)

The method of claim 15, wherein the second blocks are used for data coalescence and garbage collection.

### Claim 18 (depends on 16)

The method of claim 16 wherein the first blocks and the second blocks are dynamically assigned to each group on demand according to their configuration.
