Memory Node Error Correction via Resiliency Groups

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current memory network systems lack scalability, resilience, and flexibility in handling large data applications, leading to potential single points of failure and inefficient error correction mechanisms.

Innovation Solution

A memory network architecture with scalable, high-performance memory nodes connected via high-speed interconnects, utilizing a decoupled memory interface and flexible resiliency groups for error detection and correction, allowing for distributed memory controllers and multi-dimensional stacked memory with Reed-Solomon coding for enhanced reliability.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If traditional hard drives are used for storing terabytes of data, then storage capacity is achieved, but response times become slow

Engineering Contradiction:
Improveresponse timeVSAvoidstorage capacity
Core Design Contradiction:
SpeedVSQuantity of substance

Solution Approach 1:

The system segments storage into distributed memory nodes, each handling a portion of the total data. This allows parallel access to multiple memory modules, achieving both high storage capacity and fast response times through distributed memory architecture.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent transitions from traditional sequential storage to distributed memory access across multiple nodes, adding a spatial dimension to data storage. This enables simultaneous access to multiple data locations, dramatically improving response time while maintaining terabyte-scale capacity.

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

2Productivity

If memory nodes are connected via high-speed interconnects, then bandwidth and performance are improved, but system complexity increases

Engineering Contradiction:
Improvememory bandwidthVSAvoidinterconnect architecture
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The interconnect architecture is segmented into standardized interfaces and protocols at each memory node, reducing overall system complexity. Each node handles its own data management independently, simplifying the interconnect design while maintaining high bandwidth.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The memory nodes use universal interfaces and standardized communication protocols that can be applied across all nodes. This multi-functionality approach simplifies the interconnect architecture by using common components and protocols throughout the system.

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

3Reliability

If error correction mechanisms are implemented, then reliability is improved, but processing time and energy consumption increase

Engineering Contradiction:
Improveerror detection and correctionVSAvoiderror correction processing time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The system performs error detection and correction codes (ECC) in advance during data storage, rather than during retrieval. This preliminary action ensures data integrity without adding processing time during memory access operations.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

Memory nodes perform self-diagnosis and self-correction of errors using built-in ECC mechanisms. The nodes automatically detect and correct errors without requiring external intervention, minimizing the impact on processing time and system performance.

Inventive Principle:
Principle #25Self-service

4Adaptability or versatility

If distributed memory controllers are used, then scalability and flexibility are improved, but control complexity increases

Engineering Contradiction:
Improveconfiguration flexibilityVSAvoidcontroller management
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The memory control function is segmented and distributed across multiple independent controllers, one per memory node. This segmentation provides configuration flexibility and scalability while reducing the complexity of any single controller, as each manages only its local memory.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The distributed memory controller architecture enables dynamic configuration and reconfiguration of memory nodes without affecting the entire system. Each controller can independently manage its node, providing flexibility in system configuration while maintaining manageable complexity through localization of control functions.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP2992435B1Memory node error correction
Publication Date: 2020.12.30 HEWLETT PACKARD ENTERPRISE DEV LP
  • EP2992435B1 patent drawingFigure 1
  • EP2992435B1 patent drawingFigure 2~3
  • EP2992435B1 patent drawingFigure 4

AI summary

According to an example, a resiliency group for a memory node in a memory network can provide error correction for a memory access in the memory node. The memory access may be received from a main memory controller of a processor connected to the memory network. The memory access may be executed by a memory controller of the memory node.