Analog Content-Addressable Memory for Single-Shot Range Lookup

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing computing systems rely on separate memory arrays for storing precomputed results, which increases footprint and requires multiple search and memory access steps, hindering computational performance.

Innovation Solution

Analog content addressable memory (aCAM) devices directly output binary codes by calculating decimal outputs for a predetermined function and encoding ranges of decimal inputs into aCAM cells, eliminating the need for separate memory arrays and streamlining computations into a single lookup operation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If separate memory arrays are used for storing precomputed results, then data storage capacity is improved, but device footprint increases and computational performance deteriorates

Engineering Contradiction:
Improvedata storage capacityVSAvoiddevice footprint
Core Design Contradiction:
Quantity of substanceVSDevice complexity

Solution Approach 1:

The patent merges the memory array and computation logic into a unified content addressable memory (CAM) structure. The CAM array simultaneously performs data storage and comparison operations, eliminating the need for separate memory arrays to store precomputed results. This integration reduces device footprint while maintaining data storage capacity through the inherent parallel comparison capability of CAM architecture.

Inventive Principle:
Principle #5Merging (Combining)

2Quantity of substance

If separate memory arrays are used for storing precomputed results, then data storage capacity is improved, but computational performance deteriorates due to multiple search and memory access steps

Engineering Contradiction:
Improvedata storage capacityVSAvoidcomputational performance
Core Design Contradiction:
Quantity of substanceVSProductivity

Solution Approach 1:

The patent pre-computes lookup tables (LUTs) during the programming phase and stores them directly in the CAM array. During operation, the CAM performs single-shot comparisons without requiring multiple search steps or additional memory accesses. This preliminary preparation of computation results enables instantaneous lookup operations, dramatically improving computational performance while maintaining full data storage capacity.

Inventive Principle:
Principle #10Preliminary action

3Measurement precision

If multiple search and memory access steps are used, then data accuracy is improved, but processing time increases

Engineering Contradiction:
Improvedata accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent replaces sequential mechanical access operations with parallel content-based comparison in the CAM array. Instead of sequentially accessing memory locations and comparing data, the CAM simultaneously compares the input key against all stored keys in parallel using content addressable logic. This substitution maintains data accuracy through exact matching while reducing processing time from multiple sequential steps to a single parallel operation.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS12462872B2Analog content addressable memory for general computing
Publication Date: 2025.11.04 HEWLETT PACKARD ENTERPRISE DEV LP
  • US12462872B2 patent drawing
  • US12462872B2 patent drawing
  • US12462872B2 patent drawing

AI summary

In an implementation, a device includes: a first inverter; a first sense amplifier connected to the first inverter; a first match line connected to the first sense amplifier; a first analog content addressable memory cell including: a first pull-down transistor; a second pull-down transistor, the first pull-down transistor and the second pull-down transistor being connected in series between the first match line and a reference node; a first memory circuit configured to store a first upper bound of a first range and to activate the first pull-down transistor when an analog input value is less than the first upper bound of the first range; and a second memory circuit configured to store a first lower bound of the first range and to activate the second pull-down transistor when the analog input value is greater than the first lower bound of the first range.