Ranged Content-Addressable Memory Parallel Search Logic

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional memory systems face inefficiencies in data retrieval and management, particularly in content-driven searches, as they perform serial searches for each bit-pattern, which can be slow and resource-intensive, especially when handling a range of bit patterns.

Innovation Solution

The implementation of a ranged content-addressable memory (RCAM) system, where each memory element stores two values defining a range, allowing for parallel content-driven searches, enabling rapid identification of memory elements containing a search value within those ranges.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If conventional memory systems perform serial searches for each bit-pattern, then resource consumption is reduced, but search speed deteriorates

Engineering Contradiction:
Improvesearch speedVSAvoidresource efficiency
Core Design Contradiction:
SpeedVSProductivity

Solution Approach 1:

The patent segments the search space by dividing bit patterns into multiple fields (first field, second field, third field) and organizing memory elements to store range definitions for each field. This allows the search to be broken down into manageable segments that can be processed in parallel, resolving the contradiction by enabling fast parallel search without overwhelming system resources.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces a new dimension to the search problem by storing not just single bit patterns but ranges of bit patterns in memory elements. Each memory element stores a first value representing a most significant bit pattern and a second value representing a least significant bit pattern, creating a dimensional expansion that enables parallel range-based searching.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Speed

If parallel searches are performed across all memory elements, then search speed is improved, but device complexity increases

Engineering Contradiction:
Improvesearch speedVSAvoidmemory structure complexity
Core Design Contradiction:
SpeedVSDevice complexity

Solution Approach 1:

The patent applies local quality by making different parts of the memory structure serve different functions: some memory elements store range definitions (first and second values), while others store actual data. The control logic is distributed across memory elements rather than centralized, allowing parallel operation without requiring complex centralized control, thus improving search speed while managing device complexity.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent implements preliminary action by pre-organizing memory elements to store range definitions (most significant bit pattern and least significant bit pattern) before the search operation. This pre-structuring allows the search to proceed in parallel without requiring complex runtime decision-making, resolving the contradiction between speed and complexity.

Inventive Principle:
Principle #10Preliminary action

3Adaptability or versatility

If memory elements store two values defining a range, then search capability is improved, but manufacturing complexity increases

Engineering Contradiction:
Improvesearch capabilityVSAvoidmemory element fabrication
Core Design Contradiction:
Adaptability or versatilityVSEase of manufacture

Solution Approach 1:

The patent applies universality by designing memory elements that can serve multiple functions: they can store range definitions (first and second values) for search operations, and they can also store actual data. This multi-functionality allows the same memory structure to handle both the complexity of range-based searching and standard data storage, improving search capability while simplifying manufacturing by avoiding specialized components.

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

4Productivity

If rapid parallel searching is implemented, then productivity is improved, but energy consumption increases

Engineering Contradiction:
Improvesearch throughputVSAvoidenergy consumption
Core Design Contradiction:
ProductivityVSUse of energy by moving object

Solution Approach 1:

The patent implements partial action by performing parallel searches only on the necessary fields (first, second, and third fields) rather than examining all possible bit patterns. The search is limited to the specific range defined by stored values, allowing rapid parallel searching to improve productivity without unnecessarily consuming energy on exhaustive searches of the entire memory space.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentUS11017857B2Ranged content-addressable memory
Publication Date: 2021.05.25 MICRON TECHNOLOGY INC
  • US11017857B2 patent drawing
  • US11017857B2 patent drawing
  • US11017857B2 patent drawing

AI summary

A memory device comprises a block of ranged content-addressable memory (RCAM) including multiple RCAM memory elements, wherein each RCAM memory element is accessed by content that includes two values; a search register configured to store a search value; and logic circuitry coupled to the multiple content-addressable memory elements and the search register. The logic circuitry is configured to: compare the search value of the search register to the two values of each of the multiple RCAM memory elements in parallel; and identify an RCAM memory element according to the comparison.