Double-Buffered Memory Topology for High-Capacity DIMM Channels

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current data storage systems are limited in memory capacity, restricting the amount of data they can process effectively due to their design constraints.

Innovation Solution

A double buffering memory topology configuration with a memory controller, data buffers, and dual in-line memory modules (DIMMs) that support signal integrity and data transfer rates, allowing for expanded memory capacity and efficient data operations through optimized channel topologies and timing delays.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If current memory system designs are used, then device complexity is limited, but memory capacity is restricted

Engineering Contradiction:
Improvememory capacityVSAvoidsystem design complexity
Core Design Contradiction:
Quantity of substanceVSDevice complexity

Solution Approach 1:

The memory system is segmented into multiple independent memory channels, each capable of operating autonomously. Each channel includes its own memory controller interface, data buffers, and DIMM modules, allowing the system to scale capacity by adding channels without fundamentally redesigning the entire memory architecture.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces a dual-buffer architecture where each memory channel contains two separate buffers (buffer A and buffer B) that can operate in parallel. This dimensional expansion from single-buffer to dual-buffer per channel enables doubled memory capacity and improved throughput without proportionally increasing control logic complexity.

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

2Quantity of substance

If memory capacity is increased, then data processing capability improves, but data transfer rate may be compromised

Engineering Contradiction:
Improvememory capacityVSAvoiddata transfer rate
Core Design Contradiction:
Quantity of substanceVSSpeed

Solution Approach 1:

The dual-buffer architecture enables continuous data transfer operations by allowing one buffer to be filled while the other is being read from, eliminating idle wait states. The memory controller can switch between buffers seamlessly, maintaining continuous data flow and maximizing transfer rates even as capacity increases.

Inventive Principle:
Principle #20Continuity of useful action

Solution Approach 2:

The system employs periodic buffer switching where the memory controller alternates between buffer A and buffer B in a regular pattern. This periodic action allows precharging and refreshing of one buffer while the other is actively being accessed, maintaining high data transfer rates across expanded capacity.

Inventive Principle:
Principle #19Periodic action

3Quantity of substance

If multiple memory modules are added, then memory capacity expands, but signal integrity deteriorates

Engineering Contradiction:
Improvememory capacityVSAvoidsignal integrity
Core Design Contradiction:
Quantity of substanceVSReliability

Solution Approach 1:

By segmenting the memory system into separate channels with dedicated control paths, the patent isolates signal integrity issues to individual channels rather than allowing them to propagate system-wide. Each channel's timing and signal routing can be independently optimized and trained, maintaining reliability as capacity expands through additional channels.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Data buffers serve as intermediary components between the memory controller and DIMM modules, providing signal conditioning, buffering, and timing adjustment. These intermediaries isolate the controller from direct signal degradation effects and allow for signal refresh and re-timing before data reaches the memory modules.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentUS12141081B2Training and operations with a double buffered memory topology
Publication Date: 2024.11.12 RAMBUS INC
  • US12141081B2 patent drawing
  • US12141081B2 patent drawing
  • US12141081B2 patent drawing

AI summary

System and method for training and performing operations (e.g., read and write operations) on a double buffered memory topology. In some embodiments, eight DIMMs are coupled to a single channel. The training and operations schemes are configured with timing and signaling to allow training and operations with the double buffered memory topology. In some embodiments, the double buffered memory topology includes one or more buffers on a system board (e.g., motherboard).