# US 10,768,824 B2 · Stacked memory device and a memory chip including the same

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

- Number: US 10,768,824 B2
- Document ID: US10768824B2
- Status: Granted
- Kind: B2
- Application number: 16/418,502
- Filed: 2019-05-21
- Published: 2019-09-05
- Granted: 2020-09-08
- Earliest claimed: 2016-07-26
- Assignees: Samsung Electronics Co., Ltd.
- Applicants: Samsung Electronics Co., Ltd.
- Inventors: Hak-Soo Yu; Je-Min Ryu; Reum Oh; Pavan Kumar Kasibhatla; Seok-In Hong
- Web page: https://patentgrep.com/patent/US10768824B2

## CPC

- G06F 3/061: {Improving I/O performance}
- G06F 3/0656: {Data buffering arrangements}
- G06F 3/0683: {Plurality of storage devices}
- G11C 5/02: Disposition of storage elements, e.g. in the form of a matrix array
- G11C 5/025: {Geometric lay-out considerations of storage- and peripheral-blocks in a semiconductor storage device (geometrical lay-out of the components in integrated circuits, geometrical lay-out of the components in integrated circuits H10D89/10)}
- G11C 5/04: Supports for storage elements {, e.g. memory modules}; Mounting or fixing of storage elements on such supports
- G11C 5/06: Arrangements for interconnecting storage elements electrically, e.g. by wiring
- G11C 5/063: {Voltage and signal distribution in integrated semi-conductor memory access lines, e.g. word-line, bit-line, cross-over resistance, propagation delay}
- G11C 5/066: {Means for reducing external access-lines for a semiconductor memory clip, e.g. by multiplexing at least address and data signals}
- 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

A stacked memory includes a logic semiconductor die, a plurality of memory semiconductor dies stacked with the logic semiconductor die, a plurality of through-silicon vias (TSVs) electrically connecting the logic semiconductor die and the memory semiconductor dies, a global processor disposed in the logic semiconductor die and configured to perform a global sub process corresponding to a portion of a data process, a plurality of local processors respectively disposed in the memory semiconductor dies and configured to perform local sub processes corresponding to other portions of the data process and a plurality of memory integrated circuits respectively disposed in the memory semiconductor dies and configured to store data associated with the data process.

## Claims

### Claim 1

A stacked memory device, comprising:
a logic semiconductor die;
a plurality of memory semiconductor dies stacked with the logic semiconductor die;
a plurality of through-silicon vias (TSVs) electrically connecting the logic semiconductor die and the memory semiconductor dies;
a global processor disposed in the logic semiconductor die and configured to perform a global sub process corresponding to a portion of a data process;
a plurality of local processors respectively disposed in the memory semiconductor dies and configured to perform local sub processes corresponding to other portions of the data process; and
a plurality of memory integrated circuits respectively disposed in the memory semiconductor dies and configured to store data associated with the data process wherein each of the memory integrated circuits includes a sense amplifier configured to sense data to be read from the memory integrated circuit and the local processor is separated from the memory integrated circuit on each of the memory semiconductor dies,
wherein the global processor includes a selector and an arithmetic logic unit, the selector including multiplexers configured to select and provide at least one of internal data provided from a first memory integrated circuit and external data provided from an external device, and wherein the arithmetic logic unit generates an address to access a second memory integrated circuit based on an output of the selector.

### Claim 2 (depends on 1)

The stacked memory device of claim 1, wherein the global processor and the local processors perform dispersively the data process that is associated with fields of big data, vision recognition, search engine, signal processing in an image sensor or signal processing in a display device.

### Claim 3 (depends on 1)

The stacked memory device of claim 1, wherein the global sub process corresponds to data layout transformation including data gathering, data scattering, data transposition, or data swapping and the local sub process corresponds to data reduction including data filtering or data cleaning to reduce a data size or data bits.

### Claim 4 (depends on 1)

The stacked memory device of claim 1, wherein the global sub process corresponds to data partitioning to divide frame data into data portions and the local sub process corresponds to data coding to compress the data portions.

### Claim 5 (depends on 1)

The stacked memory device of claim 1, wherein the local sub process corresponds to data decoding to decompress compressed data and the global sub process corresponds to data combining to generate frame data by combining a plurality of decompressed data portions.

### Claim 6 (depends on 1)

The stacked memory device of claim 1, wherein the global sub process corresponds to one of coarse processing and fine processing and the local sub process corresponds to the other of the coarse processing and the fine processing.

### Claim 7 (depends on 6)

The stacked memory device of claim 6, wherein the coarse processing is a process to compare small-sized data with first reference data in data/pattern matching, and the fine processing is a process to compare large-sized data with second reference data.

### Claim 8 (depends on 6)

The stacked memory device of claim 6, wherein the coarse processing is data/pattern matching to compare data with reference data, and the fine processing is a process of analyzing an attribute or kind of matched data that is provided by the coarse processing.

### Claim 9 (depends on 1)

The stacked memory device of claim 1, wherein the global processor includes:
a plurality of processing units configured to perform different processes.

### Claim 10 (depends on 9)

The stacked memory device of claim 9, wherein the global processor selects one processing unit corresponding to the global sub process among the plurality of processing units.

### Claim 11 (depends on 1)

The stacked memory device of claim 1, wherein at least one of the local processors includes:
a comparator configured to compare reference data and read data provided from its corresponding memory integrated circuit in response to a read address to detect whether the read data includes the same data/pattern as the reference data.

### Claim 12 (depends on 11)

The stacked memory device of claim 11, wherein the at least one of the local processors further includes:
a shift register configured to sequentially shift bits of the read data when a bit number of the read data is larger than a bit number of the reference data.

### Claim 13 (depends on 11)

The stacked memory device of claim 11, wherein the at least one of the local processors further includes:
an address generator configured to generate the read address that is sequentially increasing or decreasing.

### Claim 14 (depends on 13)

The stacked memory device of claim 13, wherein the address generator determines a range of the data/pattern based on a start address and an end address provided from the global processor.

### Claim 15

A stacked memory device, comprising:
a logic semiconductor die including a global processor configured to perform a global sub process corresponding to a portion of a data process; and
a plurality of memory semiconductor dies stacked vertically, the memory semiconductor dies including a plurality of local processors configured to perform local sub processes corresponding to other portions of the data process and a plurality of memory integrated circuits configured to store data associated with the data process, wherein each of the memory integrated circuits includes a sense amplifier configured to sense data to be read from the memory integrated circuit and the local processor is separated from the memory integrated circuit on each of the memory semiconductor dies,
wherein the global processor includes a selector and an arithmetic logic unit, the selector including multiplexers configured to select and provide at least one of internal data provided from a first memory integrated circuit and external data provided from an external device, and wherein the arithmetic logic unit generates an address to access a second memory integrated circuit based on an output of the selector.

### Claim 16 (depends on 15)

The stacked memory device of claim 15, wherein the global processor and the local processors perform dispersively the data process that is associated with fields of big data, vision recognition, search engine, signal processing in an image sensor or signal processing in a display device.

### Claim 17 (depends on 15)

The stacked memory device of claim 15, wherein the global processor includes:
a plurality of processing units configured to perform different processes.

### Claim 18 (depends on 15)

The stacked memory device of claim 15, wherein at least one of the local processors includes:
a comparator configured to compare reference data and read data provided from its corresponding memory integrated circuit in response to a read address to detect whether the read data includes the same data/pattern as the reference data.

### Claim 19

A memory device, comprising:
a logic semiconductor die including a global processor configured to perform a global sub process corresponding to a first portion of a data process of an external device; and
a plurality of memory semiconductor dies, wherein the memory semiconductor dies are stacked with respect to each other and a first memory semiconductor die includes a first local processor configured to perform a local sub process corresponding to a second portion of the data process and a memory integrated circuit configured to store data associated with the data process, wherein the memory integrated circuit includes a sense amplifier configured to sense data to be read from the memory integrated circuit and the first local processor is separated from the memory integrated circuit on the first memory semiconductor die,
wherein the global processor includes a selector and an arithmetic logic unit, the selector including multiplexes configured to select and provide at least one of internal data provided from the memory integrated circuit and external data provided from the external device, and wherein the arithmetic logic unit generates an address to access the memory integrated circuit based on an output of the selector.

### Claim 20 (depends on 19)

The memory device of claim 19, wherein the global processor and the first local processor perform dispersively the data process that is associated with fields of big data, vision recognition, search engine, signal processing in an image sensor or signal processing in a display device.