# US 2019/0317691 A1 · Memory system and method for controlling nonvolatile memory

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

- Number: US 2019/0317691 A1
- Document ID: US20190317691A1
- Status: Published application
- Kind: A1
- Application number: 16/126,038
- Filed: 2018-09-10
- Published: 2019-10-17
- Earliest claimed: 2018-04-16
- Assignees: Toshiba Memory Corporation
- Applicants: Toshiba Memory Corporation
- Inventors: Shinichi KANNO
- Web page: https://patentgrep.com/patent/US20190317691A1

## CPC

- G06F 3/0656: {Data buffering arrangements}
- G06F 3/0604: {Improving or facilitating administration, e.g. storage management}
- G06F 3/061: {Improving I/O performance}
- G06F 3/0659: {Command handling arrangements, e.g. command buffers, queues, command scheduling}
- G06F 3/0679: {Non-volatile semiconductor memory device, e.g. flash memory, one time programmable memory [OTP]}
- G06F 12/0246: {in block erasable memory, e.g. flash memory}
- G11C 11/5628: {Programming or writing circuits; Data input circuits}
- G11C 11/5642: {Sensing or reading circuits; Data output circuits}
- G11C 16/0483: {comprising cells having several storage transistors connected in series}
- G11C 16/10: Programming or data input circuits
- G11C 16/26: Sensing or reading circuits; Data output circuits
- G06F 2212/7203: Temporary buffering, e.g. using volatile buffer or dedicated buffer blocks
- Y02D 10/00: Energy efficient computing, e.g. low power processors, power management or thermal management
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G11C: STATIC STORES (semiconductor memory devices H10B)
- Y02D: CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE

## Abstract

According to one embodiment, a memory system stores write data received from a host to a shared write buffer for write destination blocks, acquires first write data for plural pages from the shared write buffer, and writes the first write data to a first write destination block by a first-step write operation. When receiving write data from the host in a state in which an empty region does not exist in the shared write buffer, the memory system discards write data in the shared write buffer in which the first-step write operation has been finished. In a case where the first write data do not exist in the shared write buffer when a second-step write operation of the first write data is to be executed, the memory system transmits a request to acquire the first write data to the host.

## Claims

### Claim 1

A memory system connectable to a host, comprising:
a nonvolatile memory including a plurality of blocks; and
a controller electrically connected to the nonvolatile memory and configured to execute a multi-step write operation including at least a first-step write operation and a second-step write operation, the multi-step write operation transferring same data for plural pages to the nonvolatile memory at plural times, wherein
the controller is configured to:
receive, from the host, write data stored in any one of a plurality of write buffers on a memory included in the host, which correspond to a plurality of write destination blocks allocated from the plurality of blocks, and store the received write data in a shared write buffer shared by the write destination blocks, every time the controller receives a write request from the host;
acquire first write data for plural pages which are to be written to a first write destination block in the write destination blocks, from the shared write buffer, and write the first write data to the first write destination block by the first-step write operation;
when receiving write data from the host in a state in which an empty region does not exist in the shared write buffer, discard write data in the shared write buffer in which the first-step write operation has been finished to secure the empty region in the shared write buffer; and
in a case where the first write data do not exist in the shared write buffer when the second-step write operation of the first write data is to be executed, acquire the first write data from the host by transmitting a request to acquire the first write data to the host.

### Claim 2 (depends on 1)

The memory system of claim 1, wherein
in a case where the first write data exist in the shared write buffer when the second-step write operation of the first write data is to be executed, the controller acquires the first write data from the shared write buffer and writes the first write data to the first write destination block by the second-step write operation.

### Claim 3 (depends on 2)

The memory system of claim 2, wherein
the controller is configured to discard the first write data in the shared write buffer when the multi-step write operation of the first write data is finished.

### Claim 4 (depends on 1)

The memory system of claim 1, wherein
when receiving write data from the host in a state in which an empty region does not exist in the shared write buffer, the controller discards oldest write data, among write data in the shared write buffer in which the first-step write operation has been finished.

### Claim 5 (depends on 1)

The memory system of claim 1, wherein
each of the received write requests includes a block address designating a write destination block, and
the controller is configured to acquire write data for plural pages associated with one or more write requests designating the first write destination block, as the first write data, from the shared write buffer.

### Claim 6 (depends on 1)

The memory system of claim 1, wherein
the shared write buffer is implemented by a random-access memory included in the memory system.

### Claim 7 (depends on 1)

The memory system of claim 1, wherein
each of the write destination blocks is implemented by a block to which data of plural bits per memory cell are written, and
the shared write buffer is implemented by one or more single-level cell blocks to which data of 1 bit per memory cell are written.

### Claim 8

A memory system connectable to a host, comprising:
a nonvolatile memory including a plurality of blocks; and
a controller electrically connected to the nonvolatile memory and configured to, by a multi-step write operation including at least a first-step write operation including a transfer of data for plural pages to the nonvolatile memory and a second-step write operation including a transfer of the data to the nonvolatile memory, write the data to plural memory cells connected to a same word line in a write destination block of the nonvolatile memory, wherein
the controller is configured to:
receive, from the host, write data stored in any one of a plurality of write buffers on a memory included in the host, which correspond to a plurality of write destination blocks allocated from the plurality of blocks, and store the received write data in a shared write buffer shared by the write destination blocks, every time the controller receives a write request including a block address designating any one of the write destination blocks, from the host;
acquire first write data for plural pages associated with one or more write requests designating a first write destination block in the write destination blocks, from the shared write buffer, and write the first write data to first memory cells connected to one write target word line in the first write destination block, by the first-step write operation;
when receiving write data from the host in a state in which an empty region does not exist in the shared write buffer, discard write data in the shared write buffer in which the first-step write operation has been finished to secure the empty region in the shared write buffer,
in a case where the first write data exist in the shared write buffer when the second-step write operation of the first write data is to be executed, acquire the first write data from the shared write buffer and write the first write data to the first memory cells of the first write destination block, by the second-step write operation; and
in a case where the first write data do not exist in the shared write buffer when the second-step write operation of the first write data is to be executed, acquire the first write data from the host by transmitting a request to acquire the first write data to the host and write the first write data to the first memory cells of the first write destination block, by the second-step write operation.

### Claim 9 (depends on 8)

The memory system of claim 8, wherein
when receiving write data from the host in a state in which an empty region does not exist in the shared write buffer, the controller discards oldest write data, among write data in the shared write buffer in which the first-step write operation has been finished.

### Claim 10 (depends on 8)

The memory system of claim 8, wherein
the controller is configured to discard the first write data in the shared write buffer when the multi-step write operation of the first write data has been finished.

### Claim 11

A method of controlling a nonvolatile memory including a plurality of blocks, and executing a multi-step write operation including at least a first-step write operation and a second-step write operation, the multi-step write operation transferring same data for plural pages to the nonvolatile memory at plural times,
the method comprising:
every time of receiving a write request from the host, executing an operation of receiving, from the host, write data stored in any one of a plurality of write buffers on a memory included in the host, which correspond to a plurality of write destination blocks allocated from the plurality of blocks, and an operation of storing the received write data in a shared write buffer shared by the write destination blocks;
acquiring first write data for plural pages which are to be written to a first write destination block in the write destination blocks, from the shared write buffer, and writing the first write data to the first write destination block by the first-step write operation;
when receiving the write data from the host in a state in which an empty region does not exist in the shared write buffer, discarding write data in the shared write buffer in which the first-step write operation has been finished to secure the empty region in the shared write buffer; and
in a case where the first write data do not exist in the shared write buffer when the second-step write operation of the first write data is to be executed, acquiring the first write data from the host by transmitting a request to acquire the first write data to the host.

### Claim 12 (depends on 11)

The method of claim 11, further comprising:
in a case where the first write data exist in the shared write buffer when the second-step write operation of the first write data is to be executed, acquiring the first write data from the shared write buffer and writing the first write data to the first write destination block by the second-step write operation.

### Claim 13 (depends on 11)

The method of claim 11, further comprising:
discarding the first write data in the shared write buffer when the multi-step write operation of the first write data is finished.

### Claim 14 (depends on 11)

The method of claim 11, wherein
the discarded write data is oldest write data among the write data in the shared write buffer in which the first-step write operation has been finished.

### Claim 15 (depends on 11)

The method of claim 11, wherein
each of the received write requests includes a block address designating a write destination block, and
write data for plural pages associated with one or more write requests designating the first write destination block are acquired from the shared write buffer as the first write data.
