Raid storage device, host, and raid system

Patent metadata

Field	Value
Publication number	US-2022011979-A1
Application number	US-202117351506-A
Country	US
Kind code	A1
Filing date	Jun 18, 2021
Priority date	Jul 10, 2020
Publication date	Jan 13, 2022
Grant date	—

Abstract

A redundant array of independent disks (RAID) storage device including; a memory device including first memory devices configured to store at least one of data chunks and corresponding parity (data chunks/parity) and a second memory device configured to serve as a spare memory region, and a RAID controller including a RAID internal memory configured to store a count table and configured to control performing of a rebuild operation in response to a command received from a host, wherein upon identification of a failed first memory device, the RAID controller accesses used regions of non-failed first memory devices based on the count table and rebuilds data of the failed first memory device using the second memory device.

First claim

1 . A redundant array of independent disks (RAID) storage device comprising:
a memory device including first memory devices configured to store at least one of data chunks and corresponding parity (data chunks/parity) and a second memory device configured to serve as a spare memory region; and a RAID controller including a RAID internal memory configured to store a count table and configured to control performing of a rebuild operation in response to a command received from a host, wherein upon identification of a failed first memory device, the RAID controller accesses used regions of non-failed first memory devices based on the count table and rebuilds data of the failed first memory device using the second memory device.

2 . The RAID storage device of claim 1 , wherein the count table stores count table information indicating used regions in at least one of the first memory devices.

3 . The RAID storage device of claim 2 , wherein in response to a RAID operation on the first memory devices, the RAID controller generates a hash value by performing hash modulation based on a logical block address (LBA) at which at least one of the data chunks or parity are stored, and stores a count in a bucket corresponding to the hash value in the count table.

4 . The RAID storage device of claim 3 , wherein when the RAID operation is a write operation, the RAID controller increases a count corresponding to a region of a first memory device to which the data chunks/parity are written.

5 . The RAID storage device of claim 3 , wherein when the RAID operation is a discard operation, the RAID controller reduces a count corresponding to a region of a first memory device from which the data chunks/parity are discarded.

6 . The RAID storage device of claim 1 , wherein the RAID controller does not perform the rebuild operation when count values of all buckets in the count table are zero.

7 . A host device connected to nonvolatile memory devices including first memory devices configured to store at least one of data chunks and corresponding parity (data chunks/parity), and a second memory device configured to serve as a spare memory region, the host device comprising:
a host internal memory configured to store a count table indicating used regions of each of the first memory devices; and a host processing unit configured to control a redundant array of independent disks (RAID) operation on the first memory devices and update the count table in response to the RAID operation, and further configured to control a rebuild operation on a failed first memory device among the first memory devices using the second memory device.

8 . The host device of claim 7 , wherein the host processing unit generates a hash value by performing hash modulation based on an LBA at which at data chunks/parity are stored in non-failed first memory devices, and stores a count in a bucket corresponding to the hash value in the count table.

9 . The host device of claim 8 , wherein when the RAID operation is a write operation, the host processing unit increases in the count table a count corresponding to a first memory device region to which the data chunks/parity are written.

10 . The host device of claim 8 , wherein when the RAID operation is a discard operation, the host processing unit decreases in the count table a count corresponding to a first memory device region from which the data chunks/parity are discarded.

11 . The host device of claim 8 , wherein the host processing unit begins the rebuild operation when the failed first memory device is identified, checks used regions of the non-failed first memory devices based on the count table, performs an XOR operation by reading data from the used regions of the non-failed first memory devices, and writes a result of the XOR operation to a region of the second memory device.

12 . The host device of claim 11 , wherein the host processing unit updates the count table based, in part, on the used region of the second memory device.

13 . The host device of claim 11 , wherein when count values of all buckets of the count table are zero, the host processing unit ends the rebuild operation without performing the XOR operation.

14 . A redundant array of independent disks (RAID) system comprising:
a host and solid state drives connected to the host, wherein each one of the solid state drives includes;
first memory devices configured as a RAID and further configured to store at least one of data and corresponding parity (data/parity),
a second memory device configured to serve as a spare memory device, and
a memory controller configured to performs data access operations on the first memory device and the second memory device,
the RAID system storing a count table indicating used regions for each of the first memory devices, and upon identifying a failed first memory device, the RAID system accesses data/parity in only used regions of non-failed first memory devices by referencing the count table to perform a rebuild operation on the failed first device using the second memory device.

15 . The RAID system of claim 14 , wherein the count table includes buckets indicating positional information for used regions and unused regions for each of the first memory devices and the second memory device.

16 . The RAID system of claim 15 , wherein when data/parity is written to one of the first memory devices, the count table is updated to increase a count in a bucket corresponding to a position of the first memory device to which the data/parity is written.

17 . The RAID system of claim 15 , wherein when the data/parity is discarded from one of the first memory devices, the count table is updated to decrease a count in a bucket corresponding to a position of the first memory region from which the data/parity is discarded.

18 . The RAID system of claim 15 , wherein when the rebuild operation is performed, the count table changes a count in a bucket corresponding to a position of a used region in each of the non-failed first memory devices and the second memory device.

19 . The RAID system of claim 15 , wherein the RAID system generates a hash value performing hash modulation based on a logical block address (LBA) of a used region for each of the non-failed first memory devices, and stores a changed count in each bucket at a position corresponding to the hash value in the count table.

20 . The RAID system of claim 14 , wherein the host comprises a host internal memory configured to store the count table that is updated whenever data/parity is written to or discarded from the first memory devices.

21 - 27 . (canceled)

Assignees

• Samsung Electronics Co Ltd

Inventors

• Lim Jae Hwan
• Lim Seung-Woo
• Kim Sung-Wook
• Kim So-Geum
• Kim Jae Eun
• You Dae Hun
• Jun Walter

Classifications

• G06F3/064

  Management of blocks · CPC title

• G06F3/0658

  Controller construction arrangements · CPC title

• G06F3/0689

  Disk arrays, e.g. RAID, JBOD · CPC title

• G06F11/1092 Primary

  Rebuilding, e.g. when physically replacing a failing disk · CPC title

• G06F3/0611 Primary

  in relation to response time · CPC title