In-Memory Multiplication Using Sensing Circuitry

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing memory devices require significant power and time to perform multiplication operations, as they often need to transfer data between memory arrays and processing resources via buses, which can be inefficient and power-intensive.

Innovation Solution

The solution involves performing multiplication operations directly within the memory array using sensing circuitry, eliminating the need to transfer data via input/output lines by utilizing memory cells coupled to access and sense lines, allowing for parallel processing and reduced power consumption.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Power

If data is transferred between memory arrays and processing resources via buses, then multiplication operations can be performed, but power consumption increases and processing time increases

Engineering Contradiction:
Improvepower consumptionVSAvoidprocessing efficiency
Core Design Contradiction:
PowerVSProductivity

Solution Approach 1:

The patent merges the memory array and processing functions by enabling multiplication operations to be performed directly within the memory array using sensing circuitry. This integration eliminates the need for separate processing resources and data transfer buses, thereby reducing power consumption while maintaining or improving processing efficiency through in-situ computation.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The sensing circuitry in the memory array is designed to perform multiple functions: traditional data sensing/readout and multiplication operations. This multi-functionality allows the same hardware infrastructure to handle both storage and computation tasks, reducing the need for additional processing components and minimizing data transfer requirements.

Inventive Principle:
Principle #6Universality (Multi-functionality)

2Speed

If data is transferred between memory arrays and processing resources via buses, then multiplication operations can be performed, but computational time increases

Engineering Contradiction:
Improvecomputational speedVSAvoiddata transfer time
Core Design Contradiction:
SpeedVSLoss of time

Solution Approach 1:

By combining memory and processing functions within the same array structure, the patent eliminates data transfer time between separate components. Multiplication operations are executed in-situ using the sensing circuitry, allowing computational speed to improve while eliminating the time loss associated with data transfer via buses.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The sensing circuitry is pre-configured and positioned within the memory array to perform multiplication operations on data as it is being sensed or stored. This preliminary action approach allows computations to be performed immediately on the data without requiring subsequent transfer to external processing units, thereby reducing computational time and eliminating data transfer delays.

Inventive Principle:
Principle #10Preliminary action

3Productivity

If multiplication operations are performed within the memory array using sensing circuitry, then power consumption decreases and processing efficiency improves, but device complexity increases

Engineering Contradiction:
Improveprocessing efficiencyVSAvoidcircuit complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The sensing circuitry is designed to perform both traditional sensing functions and multiplication operations, reducing the need for separate processing components. This multi-functionality approach improves processing efficiency while minimizing the increase in device complexity by reusing existing infrastructure for multiple purposes.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The memory array's sensing circuitry is utilized to perform multiplication operations on the data it already handles during normal read operations. This self-service approach allows the memory system to perform computations independently without requiring additional external processing resources, thereby improving processing efficiency while avoiding significant increases in overall device complexity.

Inventive Principle:
Principle #25Self-service

Data Source

PatentUS10713011B2Multiplication operations in memory
Publication Date: 2020.07.14 MICRON TECHNOLOGY INC
  • US10713011B2 patent drawing
  • US10713011B2 patent drawing
  • US10713011B2 patent drawing

AI summary

Examples of the present disclosure provide apparatuses and methods for performing multiplication operations in a memory. An example method comprises performing a multiplication operation on a first element stored in a group of memory cells coupled to a first access line and a number of sense lines of a memory array and a second element stored in a group of memory cells coupled to a second access line and the number of sense lines of the memory array. The method can include a number operations performed without transferring data via an input/output (I/O) line.