Selective Memory Mirroring for Virtual Machine Reliability

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current memory redundancy techniques in information handling systems require mirroring or redundancy for the entire system memory, significantly reducing usable memory and being cost-inefficient, especially since memory is an expensive component with a low failure rate.

Innovation Solution

A method for selectively mirroring only a portion of system memory, using a memory controller to manage and report mirrored memory parameters to the operating system, allowing virtualization software to execute and protect critical data and programs from hardware failures while maintaining a larger portion of usable memory.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If memory redundancy techniques (mirroring, RAID) are applied to entire system memory, then reliability is improved, but usable memory quantity is significantly reduced

Engineering Contradiction:
Improvememory error protectionVSAvoidusable system memory
Core Design Contradiction:
ReliabilityVSQuantity of substance

Solution Approach 1:

The patent divides system memory into multiple regions and applies redundancy selectively to only those regions containing critical data or code, rather than mirroring the entire memory space. This segmentation allows reliability protection where needed while preserving usable memory capacity in non-critical regions.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements different memory configurations in different locations - critical regions receive redundancy protection while non-critical regions use standard single-memory configurations. This local quality approach ensures reliability where necessary without unnecessarily reducing overall usable memory.

Inventive Principle:
Principle #3Local quality

2Reliability

If memory redundancy techniques are applied to entire system memory, then reliability is improved, but cost increases due to memory being an expensive component

Engineering Contradiction:
Improvememory error protectionVSAvoidmemory cost
Core Design Contradiction:
ReliabilityVSLoss of energy

Solution Approach 1:

The patent segments memory into critical and non-critical regions, applying expensive redundancy techniques only to critical regions. This reduces the total amount of redundant memory needed, thereby lowering the overall cost while maintaining protection for essential system components.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies redundancy partially - only to the extent necessary for system reliability. By protecting only critical memory regions rather than all memory, the system achieves adequate protection without the excessive cost of full-memory redundancy.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentUS8151138B2Redundant memory architecture management methods and systems
Publication Date: 2012.04.03 DELL PROD LP
  • US8151138B2 patent drawing
  • US8151138B2 patent drawing
  • US8151138B2 patent drawing

AI summary

A method for managing a redundant memory architecture for an information handling system (IHS) may include providing redundant memory for a portion of system memory wherein the redundant memory is configured to mirror data stored in the portion of system memory. The method may further include reporting a parameter associated with the portion of system memory to an operating system of the IHS and executing a program from the portion of system memory.