DRAM Memory Layout for Compact Row-Hammer Protection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The increasing capacity and reduced geometry of dynamic random access memories (DRAM) exacerbate the row-hammering effect, making existing preventive refresh methods less effective and resource-intensive.

Innovation Solution

A memory device architecture that divides DRAM circuits into a directly protected part and an indirectly protected part, using two algorithms to manage the row-hammering effect, with a main table stored in the directly protected part to optimize resource usage.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If the memory device capacity is increased, then the storage capability is improved, but the size of the protection tables must be increased which consumes more resources

Engineering Contradiction:
Improvememory capacityVSAvoidresource consumption
Core Design Contradiction:
Quantity of substanceVSLoss of substance

Solution Approach 1:

The patent divides the memory device into multiple sub-banks, each with its own independent protection table. This segmentation allows the total protection resources to be distributed across sub-banks rather than requiring a single large table, thereby managing large-capacity memory while controlling resource consumption at each sub-bank level.

Inventive Principle:
Principle #1Segmentation

2Manufacturing precision

If the geometry is reduced to increase density, then the manufacturing precision is improved, but the scope of row-hammering effect increases affecting more rows

Engineering Contradiction:
Improvegeometry scalingVSAvoidrow-hammering effect scope
Core Design Contradiction:
Manufacturing precisionVSObject-generated harmful factors

Solution Approach 1:

By dividing the memory into sub-banks with independent protection mechanisms, the patent limits the propagation of row-hammering effects within each sub-bank. This segmentation contains the harmful effects to smaller scopes, preventing them from affecting the entire memory device even when geometry is reduced and density is increased.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces intermediate protection tables at the sub-bank level that act as mediators between the physical memory rows and the control logic. These intermediate structures enable independent management of row activations and prevent cascading effects across the entire memory device.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Reliability

If the critical hammering value is lowered to protect against finer patterns, then the reliability is improved, but the tables must be updated more frequently increasing resource usage

Engineering Contradiction:
Improveprotection effectivenessVSAvoidresource consumption
Core Design Contradiction:
ReliabilityVSLoss of substance

Solution Approach 1:

The patent distributes the burden of frequent table updates across multiple sub-banks, each with its own protection table. This segmentation allows the system to maintain low critical hammering values for reliability while spreading the resource consumption of frequent updates across independent sub-banks, preventing any single sub-bank from being overwhelmed.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentUS20250292820A1Memory device provided with dram memory circuits arranged in such a way as to minimize the size of a memory block allowing management of the row-hammering effect
Publication Date: 2025.09.18 QUALCOMM TECHNOLOGIES INC
  • US20250292820A1 patent drawing
  • US20250292820A1 patent drawing
  • US20250292820A1 patent drawing

AI summary

The invention relates to a memory device comprising:—DRAM memory circuits (100), the total capacity of which is divided into a first part (102) and a second part (103) larger than the first part (102);—a control circuit configured to access the memory circuits, the control circuit comprising:—a first block (201) configured to execute a first algorithm (201A) intended to protect the first part (102) from a row-hammering effect;—a second block (202) configured to execute a second algorithm (202A) intended to protect the second part (103) from a row-hammering effect that may occur, the second algorithm (202A) using a main table stored in the first part (102).