# US 2024/0004667 A1 · Parallel processing device

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

- Number: US 2024/0004667 A1
- Document ID: US20240004667A1
- Status: Published application
- Kind: A1
- Application number: 18/369,532
- Filed: 2023-09-18
- Published: 2024-01-04
- Earliest claimed: 2021-03-17
- Assignees: MORUMI Co., Ltd.
- Applicants: Morumi Co., Ltd.
- Inventors: Tae Hyoung KIM
- Web page: https://patentgrep.com/patent/US20240004667A1

## CPC

- G06F 9/3885: {using a plurality of independent parallel functional units}
- G06F 9/3895: {for complex operations, e.g. multidimensional or interleaved address generators, macros}
- G06F 7/50: Adding; Subtracting (G06F7/483 - G06F7/491, G06F7/544 - G06F7/556 take precedence)
- G06F 7/5443: {Sum of products (for applications thereof, see the relevant places, e.g. G06F17/10, H03H17/00)}
- G06F 9/38: Concurrent instruction execution, e.g. pipeline or look ahead
- G06F 9/5027: {the resource being a machine, e.g. CPUs, Servers, Terminals}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

A parallel processing device includes a preprocessing part comprising preprocessing units and a main processing part comprising summers. Each of the preprocessing units comprises a selection operator and a shift operator. The selection operator is configured to operate when corresponding one of the preprocessing units operates in a summing mode, and configured to perform a function of transmitting corresponding some first signals of first signals to corresponding one of the summers according to bits of corresponding one of second signals respectively. The shift operator is configured to operate when the corresponding preprocessing unit operates in a multiplication mode, and configured to transmit signals resulting from shifting corresponding one of the first signals to the corresponding summer according to the bits respectively.

## Claims

### Claim 1

A parallel processing device configured to receive first to N-th inputs ({X1, Y1}, {X2, Y2}, . . . , and {XN, YN}) and output first to N-th outputs (M1, M2, . . . , and MN), wherein the first to N-th inputs ({X1, Y1}, {X2, Y2}, . . . , and {XN, YN}) include a plurality of first signals (X1, X2, . . . , and XN) and a plurality of second signals (Y1, Y2, . . . , and YN) and the N is a natural number of 4 or more, the parallel processing device comprising:
a preprocessing part configured to transmit an i-th first signal (Xi) to each summer (SUM1, SUM2, . . . , and SUMN) according to i-th bits (Y1[i], Y2[i], . . . , and YN[i]) of the plurality of second signals (Y1, Y2, . . . , and YN) when operating in a summing mode, or respectively transmit signals ((X1<<(i-1)), (X2<<(i-1)), . . . , and (XN<<(i-1))) resulting from shifting the first signals (X1, X2, . . . , and XN) by (i-1) bits to the summers (SUM1, SUM2, . . . , and SUMN) according to the i-th bits (Y1[i], Y2[i], . . . , and YN[i]) of the second signals (Y1, Y2, . . . , and YN) when operating in a multiplication mode, wherein the i is a natural number that is equal to or greater than 1 and is equal to or less than the N; and
a main processing part comprising the summers (SUM1, SUM2, . . . , and SUMN) configured to output the first to N-th outputs (M1, M2, . . . , and MN), wherein each of the summers (SUM1, SUM2, . . . , and SUMN) adds the transmitted signals.

### Claim 2 (depends on 1)

The parallel processing device of claim 1, wherein the preprocessing part comprises:
a selection operator configured to transmit each of the first signals (X1, X2, . . . , and XN) to an i-th summer (SUMi) according to bits (Yi[1], Yi[2], . . . , and Yi[N]) of an i-th second signal (Yi) when operating in the summing mode; and
a shift operator configured to transmit signals ((Xi<<O), (Xi<<1), . . . , and (Xi<<(N-1))) resulting from shifting the i-th first signal (Xi) by 0, 1, . . . , and (N-1) bits to the i-th summer (SUMi) according to the bits (Yi[i], Yi[2], . . . , and Yi[N]) of the i-th second signal (Yi) when operating in a shift mode.

### Claim 3 (depends on 1)

The parallel processing device of claim 1, further comprising:
a delay part configured to delay the outputs (M1, M2, . . . , and MN) according to a clock signal and output results of delay; and
a selection part configured to output, as the first signals (X1, X2, . . . , and XN), signals selected among signals (R1, R2, . . . , and RN) transmitted from a memory and signals (D1, D2, . . . , and DN) output from the delay part, according to input control signals respectively.

### Claim 4 (depends on 1)

The parallel processing device of claim 1, wherein when the preprocessing part operates in the summing mode, an i-th output (Mi) corresponds to a sum of signals selected among the first signals (X1, X2, . . . , and XN) according to bits (Yi[1], Yi[2], . . . , and Yi[N]) of an i-th second signal (Yi).

### Claim 5 (depends on 1)

The parallel processing device of claim 1, wherein when the preprocessing part operates in the multiplication mode, an i-th output (Mi) corresponds to a product (Xi * Yi) of the i-th first signal (Xi) and an i-th second signal (Yi).

### Claim 6

A parallel processing device, comprising:
a preprocessing part comprising preprocessing units; and
a main processing part comprising summers,
wherein each of the preprocessing units comprises a selection operator and a shift operator,
the selection operator is configured to operate when corresponding one of the preprocessing units operates in a summing mode, and configured to perform a function of transmitting first signals to corresponding one of the summers according to bits of corresponding one of second signals respectively, and
the shift operator is configured to operate when the corresponding preprocessing unit operates in a multiplication mode, and configured to transmit signals resulting from shifting corresponding one of the first signals to the corresponding summer according to the bits respectively.

### Claim 7 (depends on 6)

The parallel processing device of claim 6, wherein the selection operator comprises demultiplexers, and the demultiplexers are configured to respectively output signals selected among the first signals and 0 according to the bits.

### Claim 8 (depends on 6)

The parallel processing device of claim 6, wherein the shift operator comprises shift units, and the shift units are configured to respectively output signals selected among the signals resulting from shifting and 0, according to the bits.

### Claim 9 (depends on 6)

The parallel processing device of claim 6, further comprising:
a delay part configured to delay outputs of the summers according to a clock signal and output results of delay; and
a selection part configured to respectively output, as the first signals, signals selected among signals transmitted from a memory and signals output from the delay part, according to input control signals respectively.

### Claim 10 (depends on 6)

The parallel processing device of claim 6, wherein when the corresponding preprocessing unit operates in the summing mode, an output of the corresponding summer corresponds to a sum of first signals selected among the first signals according to the bits of the corresponding second signal.

### Claim 11 (depends on 6)

The parallel processing device of claim 6, wherein when the corresponding preprocessing unit operates in the multiplication mode, an output of the corresponding summer corresponds to a product of the corresponding first signal and the corresponding second signal.

### Claim 12 (depends on 6)

The parallel processing device of claim 6, wherein some of the preprocessing units operate in the summing mode, and simultaneously, others operate in the multiplication mode.

### Claim 13 (depends on 12)

The parallel processing device of claim 12, wherein which ones of the preprocessing units operate in the summing mode and which other ones of the preprocessing units operate in the multiplication mode change over time.

### Claim 14 (depends on 6)

The parallel processing device of claim 6, wherein the corresponding preprocessing unit operates in either the summing mode or the multiplication mode according to corresponding one of a plurality of selection signals.

### Claim 15

A parallel processing device, comprising:
a preprocessing part comprising preprocessing units; and
a main processing part comprising summers,
wherein each of the preprocessing units comprises a selection operator and a shift operator,
the selection operator is configured to operate when corresponding one of the preprocessing units operates in a summing mode, and configured to perform a function of transmitting corresponding some first signals of first signals to corresponding one of the summers according to bits of corresponding one of second signals respectively, and
the shift operator is configured to operate when the corresponding preprocessing unit operates in a multiplication mode, and configured to transmit signals resulting from shifting corresponding one of the first signals to the corresponding summer according to the bits respectively.

### Claim 16 (depends on 15)

The parallel processing device of claim 15, wherein the corresponding some first signals include the corresponding first signal and first signals adjacent to the corresponding first signal.

### Claim 17 (depends on 15)

The parallel processing device of claim 15, further comprising:
a delay part configured to delay outputs of the summers according to a clock signal and output results of delay; and
a selection part configured to output, as the first signals, signals selected among signals transmitted from a memory and signals output from the delay part, according to input control signals respectively.

### Claim 18 (depends on 15)

The parallel processing device of claim 15, wherein when the corresponding preprocessing unit operates in the summing mode, an output of the corresponding summer corresponds to a sum of signals selected among the corresponding some first signals according to the bits.

### Claim 19 (depends on 15)

The parallel processing device of claim 15, wherein when the corresponding preprocessing unit operates in the multiplication mode, an output of the corresponding summer corresponds to a product of the corresponding first signal and the corresponding second signal.

### Claim 20 (depends on 15)

The parallel processing device of claim 15, wherein some of the preprocessing units operate in the summing mode, and simultaneously, others operate in the multiplication mode.

### Claim 21 (depends on 20)

The parallel processing device of claim 20, wherein which ones of the preprocessing units operate in the summing mode and which other ones of the preprocessing units operate in the multiplication mode change over time.

### Claim 22 (depends on 15)

The parallel processing device of claim 15, wherein the corresponding preprocessing unit operates in either the summing mode or the multiplication mode according to corresponding one of a plurality of selection signals.
