Virtual Machine Memory Management via BMC-RAC Failover

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current information handling systems, particularly in data centers, face challenges with memory management due to the inefficiencies in firmware features storage and customization, leading to unnecessary memory usage and limited scalability, as well as complexities in managing devices across different hardware capabilities.

Innovation Solution

The method involves assigning memory modules between compute and storage nodes, detecting connectivity losses, and establishing secondary communication channels to transfer data, allowing for dynamic memory allocation and management, including the use of non-volatile and volatile memory modules, to ensure continuous operation of virtual machines.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If memory modules are assigned to virtual machines over a communication channel between compute and storage nodes, then memory management flexibility and scalability are improved, but system complexity and connectivity failure risks increase

Engineering Contradiction:

Improvememory management flexibilityVSAvoidsystem complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system separates memory resources into distinct modules located at storage nodes, with each virtual machine assigned specific memory modules. This segmentation allows independent management and allocation of memory resources without affecting the entire system, resolving the contradiction between flexibility and complexity.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

A management controller is introduced as an intermediary component that handles memory allocation, monitoring, and failover operations. This intermediary abstracts the complexity of distributed memory management, providing simplified control while maintaining the flexibility of modular memory assignment.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Adaptability or versatility

If memory modules are located at storage nodes rather than compute nodes, then memory resource sharing and scalability are improved, but access speed and connectivity reliability deteriorate

Engineering Contradiction:

Improveresource sharing capabilityVSAvoidmemory access speed

Core Design Contradiction:

Adaptability or versatilityVSSpeed

Solution Approach 1:

The system performs preliminary actions by assigning memory modules to virtual machines before connectivity issues occur. When connectivity is lost, pre-configured failover mechanisms immediately activate, transferring memory access to alternative locations without interruption to the virtual machine operations, thus maintaining speed despite distributed architecture.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

Redundant memory modules are pre-positioned at compute nodes as cushioning resources. When storage node connectivity fails, these pre-positioned resources immediately compensate for the loss, maintaining access speed while preserving the scalability benefits of distributed memory architecture.

Inventive Principle:

Principle #11Beforehand cushioning (Prior cushioning)

3Reliability

If connectivity loss detection and failover mechanisms are implemented, then system reliability is improved, but device complexity and management overhead increase

Engineering Contradiction:

Improvesystem reliabilityVSAvoidmanagement complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The management controller implements continuous feedback monitoring of connectivity between compute and storage nodes. When connectivity loss is detected, the controller automatically triggers failover procedures, providing reliable operation without requiring complex manual intervention or management overhead.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system performs self-service failover operations where the management controller automatically detects connectivity issues and reallocates memory resources without external intervention. This automation improves reliability while reducing management complexity by eliminating the need for manual fault response procedures.

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS20250021366A1Virtual machine memory management

Publication Date: 2025.01.16 DELL PROD LP

AI summary

Virtual machine memory management, including assigning, for each virtual machine operating at a compute node, respective first memory modules to the virtual machine operating at a storage node; accessing, by the virtual machines and over a first communication channel between the compute node and the storage node, the respective first memory modules; detecting, for a particular virtual machine, a loss of connectivity of the particular virtual machine with respective first memory modules assigned to the particular virtual machine, and in response: assigning, for the particular virtual machine, second memory modules to the particular virtual machine operating at the compute node; establishing a second communication channel between a BMC of the storage node and a RAC of the computing node; transferring, over the second communication channel, data stored at the respective first memory modules assigned to the particular virtual machine to the second memory modules assigned to the particular virtual machine.