# US 2019/0377633 A1 · Providing additional parity for non-standard sized parity data sets

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

- Number: US 2019/0377633 A1
- Document ID: US20190377633A1
- Status: Published application
- Kind: A1
- Application number: 16/212,242
- Filed: 2018-12-06
- Published: 2019-12-12
- Earliest claimed: 2018-06-11
- Assignees: Seagate Technology LLC
- Applicants: Seagate Technology LLC
- Inventors: Stephen H. Perlmutter; Steven S. Williams; Benjamin J. Scott; Andrew J. Louder; Kyumsung Lee; Robert W. Dixon
- Web page: https://patentgrep.com/patent/US20190377633A1

## CPC

- G06F 11/1068: {in sector programmable memories, e.g. flash disk (G06F11/1072 takes precedence)}
- G06F 11/1012: {using codes or arrangements adapted for a specific type of error (G06F11/1048 takes precedence)}
- G06F 11/1076: {Parity data used in redundant arrays of independent storages, e.g. in RAID systems}
- G06F 12/0246: {in block erasable memory, e.g. flash memory}
- G11C 29/52: Protection of memory contents; Detection of errors in memory contents
- H03M 13/2906: {using block codes (H03M13/2957 takes precedence)}
- G06F 2212/7205: Cleaning, compaction, garbage collection, erase control
- G06F 2212/7208: Multiple device management, e.g. distributing data over multiple flash devices
- G11C 29/74: {using duplex memories, i.e. using dual copies}
- H03M 13/1102: {Codes on graphs and decoding on graphs, e.g. low-density parity check [LDPC] codes}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G11C: STATIC STORES (semiconductor memory devices H10B)
- H03M: CODING; DECODING; CODE CONVERSION IN GENERAL (using fluidic means F15C4/00; optical analogue/digital converters G02F7/00; coding, decoding or code conversion, specially adapted for particular applications, see the relevant subclasses, e.g. G01D, G01R, G06F, G06T, G09G, G10L, G11B, G11C, H04B, H04L, H04M, H04N; ciphering or deciphering for cryptography or other purposes involving the need for secrecy G09C)

## Abstract

Apparatus and method for storing data in a non-volatile memory (NVM), such as a flash memory in a solid-state drive (SSD). In some embodiments, a distributed storage space of the NVM is defined to extend across a plural number of regions of the NVM. A non-standard parity data set is provided having a plural number of data elements greater than or equal to the plural number of regions in the storage space. The data set is written by storing a first portion of the data elements and a first parity value to the plural number of regions and a remaining portion of the data elements and a second parity value to a subset of the plural number of regions. The regions can comprise semiconductor dies in a flash memory, and the distributed storage space can be a garbage collection unit formed using one erasure block from each flash die.

## Claims

### Claim 1

A method comprising:
arranging a distributed storage space of a non-volatile memory (NVM) across a plural number of regions; and
writing a non-standard parity data set to the distributed storage space having a plural number of data elements greater than or equal to the plural number of regions by storing a first portion of the data elements and a first parity value to the plural number of regions and a remaining portion of the data elements and a second parity value to a subset of the plural number of regions.

### Claim 2 (depends on 1)

The method of claim 1, wherein the plural number of regions constitutes N regions, and the plural number of elements in the parity data set plus the first parity value constitute P elements where N and P are integers and N<P<2N.

### Claim 3 (depends on 2)

The method of claim 2, wherein the first portion of the data elements comprises P−1 data elements respectively written to each of N−1 regions, and the remaining portion of the data elements comprises P−N elements written to each of P−N regions.

### Claim 4 (depends on 1)

The method of claim 1, wherein each region comprises a semiconductor die of the NVM and the distributed storage space comprises a garbage collection unit (GCU) which spans the plural number of dies in the NVM.

### Claim 5 (depends on 4)

The method of claim 4, wherein the NVM comprises flash memory.

### Claim 6 (depends on 1)

The method of claim 1, further comprising updating a map as a data structure in a memory that identifies a location of the first parity value and a location of the second parity value in the NVM.

### Claim 7 (depends on 1)

The method of claim 1, wherein each of the data elements comprises a selected amount of user data and an inner code configured to detect and correct bit errors in the user data during a read operation, and wherein each of the first and second parity values comprise an outer code configured to reconstruct an associated number of the data elements responsive to an unavailability condition associated with a selected one of the regions.

### Claim 8 (depends on 1)

The method of claim 1, wherein the first parity value is determined responsive to an exclusive-or (XOR) combination of the first portion of the data elements, and the second parity value is determined responsive to an XOR combination of the remaining portion of the data elements.

### Claim 9 (depends on 1)

The method of claim 1, wherein the NVM is characterized as a flash memory in a solid-state drive (SSD).

### Claim 10

A method comprising:
forming a garbage collection unit (GCU) across a plural number of dies characterized as semiconductor non-volatile memory (NVM) dies;
dividing a data set into a plural number of data payload elements equal to or greater than the plural number of dies;
combining a first portion of the data payload elements to form a first parity value and distributing the first portion of the data payload elements and the first parity value across the plural number of dies;
combining a remaining second portion of the data payload elements to form a second parity value and distributing the remaining second portion of the data payload elements and the second parity value across less than the plural number of dies; and
writing map data to a memory to describe the locations of the first and second parity values among the plural number of dies.

### Claim 11 (depends on 10)

The method of claim 10, wherein the plural number of dies is characterized as N dies and the plural number of data payload elements is characterized as D data elements so that D≥N, wherein the first portion of the data payload elements comprise N−1 data payload elements and the remaining second portion of the data payload elements comprise D−(N−1) elements, and wherein the first and second parity values are written to different dies.

### Claim 12 (depends on 10)

The method of claim 10, wherein each of the data elements comprises a selected amount of user data and an inner code configured to detect and correct bit errors in the user data during a read operation, and wherein each of the first and second parity values comprise an outer code configured to reconstruct an associated number of the data elements responsive to an unavailability condition associated with a selected one of the NVM dies.

### Claim 13 (depends on 12)

The method of claim 12, wherein the first parity value is determined responsive to an exclusive-or (XOR) combination of the first portion of the data payload elements, and the second parity value is determined responsive to an XOR combination of the second portion of the data payload elements.

### Claim 14 (depends on 10)

The method of claim 10, wherein the NVM dies form a flash memory in a solid-state drive (SSD).

### Claim 15

A data storage device comprising:
a non-volatile memory (NVM) comprising semiconductor dies arranged into at least one garbage collection unit (GCU), the at least one GCU comprising an erasure block selected from each of a plural number of the dies;
a controller circuit configured to write a non-standard parity data set having a plural number data elements greater than or equal to the plural number of the dies by storing an initial set of the data elements and a first parity value to each of the plural number of the dies and by storing a remaining set of the data elements and a second parity value to a subset of the dies, the first parity value protecting the initial set of the data elements and the second parity value protecting the remaining set of the data elements.

### Claim 16 (depends on 15)

The data storage device of claim 15, characterized as a solid-state drive (SSD), wherein the semiconductor dies comprise flash memory dies.

### Claim 17 (depends on 15)

The data storage device of claim 15, wherein the controller circuit is further configured to update a map as a data structure in a memory to identify a first address in the corresponding die at which the first parity value is stored and to identify a second address in the corresponding die at which the second parity value is stored.

### Claim 18 (depends on 15)

The data storage device of claim 15, wherein each of the data elements comprises a selected amount of user data and an inner code configured to detect and correct bit errors in the user data during a read operation, and wherein each of the first and second parity values comprise an outer code configured to reconstruct an associated number of the data elements responsive to an unavailability condition associated with a selected one of the regions.

### Claim 19 (depends on 15)

The data storage device of claim 15, wherein the first parity value is determined responsive to an exclusive-or (XOR) combination of the initial set of the data elements, and the second parity value is determined responsive to an XOR combination of the remaining set of the data elements.

### Claim 20 (depends on 15)

The data storage device of claim 15, wherein the control circuit is further configured to initially establish the at least one GCU as having M dies where M=N+1, wherein the control circuit reduces the size of the at least one GCU to N dies responsive to a failure of one of the M dies, and wherein the parity data set has D data elements where D≥N.
