# US 2021/0280232 A1 · Copy data in a memory system with artificial intelligence mode

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

- Number: US 2021/0280232 A1
- Document ID: US20210280232A1
- Status: Published application
- Kind: A1
- Application number: 17/328,751
- Filed: 2021-05-24
- Published: 2021-09-09
- Earliest claimed: 2019-08-29
- Assignees: Micron Technology, Inc.
- Applicants: Micron Technology, Inc.
- Inventors: Alberto Troia
- Web page: https://patentgrep.com/patent/US20210280232A1

## CPC

- G11C 7/22: Read-write [R-W] timing or clocking circuits; Read-write [R-W] control signal generators or management
- G06F 11/1448: {Management of the data involved in backup or backup restore}
- G06F 12/0246: {in block erasable memory, e.g. flash memory}
- G06F 18/214: Generating training patterns; Bootstrap methods, e.g. bagging or boosting
- G06N 3/048: Activation functions
- G06N 3/0499: Feedforward networks
- G06N 3/08: Learning methods
- G11C 7/1006: {Data managing, e.g. manipulating data before writing or reading out, data bus switches or control circuits therefor}
- G11C 7/1048: {Data bus control circuits, e.g. precharging, presetting, equalising}
- G11C 11/409: Read-write [R-W] circuits
- G11C 11/54: using elements simulating biological cells, e.g. neuron
- G06N 3/063: using electronic means
- G11C 2207/2236: Copy
- G11C: STATIC STORES (semiconductor memory devices H10B)
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

The present disclosure includes apparatuses and methods related to copying data in a memory system with an artificial intelligence (AI) mode. An apparatus can receive a command indicating that the apparatus operate in an artificial intelligence (AI) mode, a command to perform AI operations using an AI accelerator based on a status of a number of registers, and a command to copy data between memory devices that are performing AI operations. The memory system can copy neural network data, activation function data, bias data, input data, and/or output data from a first memory device to a second memory device, such that that the first memory device can use the neural network data, activation function data, bias data, input data, and/or output data in a first AI operation and the second memory device can use the neural network data, activation function data, bias data, input data, and/or output data in a second AI operation.

## Claims

### Claim 1

An apparatus, comprising:
a first memory device coupled to a controller;
a second memory device coupled to the controller; and
the controller configured to:
receive a first command indicating that the apparatus operate in an artificial intelligence (AI) mode;
receive a second command to perform AI operations using an AI accelerator based on a status of each of a number of registers; and
receive a third command to transmit data from the first memory device to the second memory device, wherein the first memory device and the second memory device are performing the AI operations using the data.

### Claim 2 (depends on 1)

The apparatus of claim 1, further comprising the controller configured to compare results of the AI operations on the first device to results of the AI operations on the second device.

### Claim 3 (depends on 1)

The apparatus of claim 1, further comprising the controller configured to receive the third command to transmit input data for a training or inference operation executed at the second memory device.

### Claim 4 (depends on 1)

The apparatus of claim 1, further comprising the controller configured to receive the third command to transmit neural network data for a training or inference operation executed at the second memory device.

### Claim 5 (depends on 1)

The apparatus of claim 1, further comprising the controller configured to receive a fourth command to transmit the data from the first memory device to a third memory device identified as a target for the data during performance of the AI operations.

### Claim 6 (depends on 1)

The apparatus of claim 1, wherein each one of the number of registers is configured to provide an input function, an output function, a neural network function, an activation function, or any combination thereof for the AI accelerator.

### Claim 7 (depends on 1)

The apparatus of claim 1, wherein the first command, the second command, and the third command comprise a single command.

### Claim 8

A system, comprising:
a number of memory devices coupled to a controller, wherein each of the number of memory devices is configured as part of a neural network; and
the controller configured to:
send a first command to a first memory device of the number of memory device to:
read data from a first memory array in the first memory device and transmit the data to a second memory array in a second memory device of the number of memory devices; and
perform an artificial intelligence (AI) operation on the first device using the data; and
send a second command to the second memory device to:
receive the data and write the data to the memory array in the second memory device; and
perform an AI operation on the second device using the data.

### Claim 9 (depends on 8)

The system of claim 8, further comprising the controller configured to compare results of the AI operation on the first device to results of the AI operation on the second device.

### Claim 10 (depends on 8)

The system of claim 8, wherein the first command further comprises a command to execute a first training or inference operation on the first memory device.

### Claim 11 (depends on 8)

The system of claim 8, wherein the second command further comprises a command to execute a second training or inference operation on the second memory device.

### Claim 12 (depends on 8)

The system of claim 8, wherein the second command comprises a first address and a second address, the first address identifying the first memory device storing data associated with a first training or inference operation and the second address identifying the second memory device as a target for the data.

### Claim 13 (depends on 8)

The system of claim 8, wherein the controller is further configured to send the first command to read the data from the first memory array and transmit the data to a third memory array in a third memory device of the number of memory devices.

### Claim 14 (depends on 13)

The system of claim 13, further comprising the controller configured to send a third command to the third memory device to perform another AI operation on the third memory device using the data transmitted from the first memory device.

### Claim 15

A method, comprising:
receiving a command that comprises a first address identifying a first memory device of a memory system and a second address identifying a second memory device of the memory system;
transmitting data from the first memory device to the second memory device in response to the command;
enabling the second memory device to perform an artificial intelligence (AI) operation by programming a register on the second memory device to enter AI mode based at least on the first address and the second address of the command; and
performing the AI operation on the second memory device using the data transmitted from the first memory device.

### Claim 16 (depends on 15)

The method claim 15, further comprising performing another AI operation on the first memory device.

### Claim 17 (depends on 16)

The method of claim 16, further comprising comparing results of the AI operation on the second memory device to the another AI operation on the first memory device.

### Claim 18 (depends on 15)

The method of claim 15, further comprising performing the AI operation based on a status of each one of a plurality of registers.

### Claim 19 (depends on 15)

The method of claim 15, further comprising selecting the first memory device and the second memory device to transfer the data on a bus shared by the first memory device, a second memory device, and a number of additional memory devices.

### Claim 20 (depends on 16)

The method of claim 16, further comprising performing the AI operation and the another AI operation using neural network data transmitted from the first memory device to the second memory device.
