Hybrid SRAM Latch Array With Balanced Fast-Slow Read Paths

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing SRAM bit cells, particularly 6T and 8T designs, face challenges in efficiently utilizing silicon area, leading to increased costs and inefficiencies in scaling with newer technology nodes.

Innovation Solution

Implementing a hybrid SRAM design using standard cell design rules with alternating rows of high-performance and high-density bit cells, combined with multiplexer circuits and write masks, to optimize area utilization and balance performance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional 6T or 8T bit cells are used in SRAM, then the SRAM can store data, but the silicon area utilization is inefficient leading to increased costs

Engineering Contradiction:
Improvedata storage capabilityVSAvoidsilicon area
Core Design Contradiction:
ReliabilityVSArea of stationary object

Solution Approach 1:

The SRAM array is divided into alternating rows of high-performance bit cells and high-density bit cells. This segmentation allows different portions of the array to serve different functions: some rows prioritize performance while others prioritize area efficiency, resolving the contradiction between reliable data storage and silicon area utilization.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Different regions of the SRAM array are assigned different cell types based on local requirements. High-performance bit cells are placed in specific rows where performance is critical, while high-density bit cells are placed in other rows where area efficiency is prioritized. This local differentiation optimizes both data storage reliability and area utilization simultaneously.

Inventive Principle:
Principle #3Local quality

2Speed

If high-performance bit cells are used throughout the SRAM array, then data access speed is improved, but silicon area usage increases leading to higher costs

Engineering Contradiction:
Improvedata access speedVSAvoidsilicon area
Core Design Contradiction:
SpeedVSArea of stationary object

Solution Approach 1:

The array is segmented into alternating rows where odd rows contain high-performance bit cells optimized for speed, and even rows contain high-density bit cells optimized for area efficiency. This segmentation ensures that high-speed access is available where needed while maintaining overall area efficiency across the entire array.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

High-performance characteristics are applied locally to specific rows rather than uniformly across the entire array. This allows the SRAM to achieve high data access speed in critical regions while maintaining cost-effectiveness through area-efficient cells in non-critical regions.

Inventive Principle:
Principle #3Local quality

3Area of stationary object

If alternating rows of high-performance and high-density bit cells are implemented, then area utilization is optimized, but performance balance between adjacent bits must be maintained

Engineering Contradiction:
Improvesilicon area utilizationVSAvoidperformance balance
Core Design Contradiction:
Area of stationary objectVSReliability

Solution Approach 1:

The patent combines high-performance multiplexer circuits with high-density bit cells in alternating rows. This merging compensates for the lower performance of high-density cells by providing high-performance read/write path circuitry, thereby maintaining overall performance balance while achieving optimized area utilization.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

High-performance multiplexer circuits act as intermediaries between the high-density bit cells and the external interface. These multiplexers compensate for the performance limitations of high-density cells, ensuring that data access performance remains balanced across adjacent bits despite the heterogeneous cell structure.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP4334938B1Hybrid library latch array
Publication Date: 2026.02.18 ADVANCED MICRO DEVICES INC
  • EP4334938B1 patent drawingFigure 1~2
  • EP4334938B1 patent drawingFigure 3~4
  • EP4334938B1 patent drawingFigure 5

AI summary

A static random access memory (SRAM) includes fast SRAM bit cells and fast multiplexer circuits that are formed in a first row of fast cells in a hybrid standard cell architecture. Slow SRAM bit cells and slow multiplexer circuits are formed in a second row of slow cells. The slow multiplexer circuits provide a column output for the fast SRAM bit cells and the fast multiplexer circuits provide a column output for the slow SRAM bit cells. Thus, one SRAM column has fast bit cells and slow multiplexer stages while the adjacent SRAM column has slow bit cells and fast multiplexer stages to thereby provide an improved performance balance when reading the SRAM.