Virtual Machine Failure Detection via Intermediary Agency

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In virtualization systems, when a virtual machine directly uses a physical device, the virtualization device manager cannot detect failures in the physical device, leading to undetected issues and potential service disruptions.

Innovation Solution

Incorporating a virtual machine failure management system that detects failures in physical devices used by virtual machines and notifies an information processing device failure management system, allowing for indirect detection and management of physical device failures.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If a virtual machine directly uses a physical device, then I/O performance and communication speed are improved, but the virtualization device manager cannot detect failures in the physical device

Engineering Contradiction:
ImproveI/O performanceVSAvoidfailure detection capability
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent introduces a failure detection agency as an intermediary component that runs within the virtual machine environment. This agency acts as a mediator between the physical device and the virtualization device manager, enabling failure detection without requiring the physical device to be directly accessible to the manager. The agency collects failure information from the physical device through the virtual machine's direct access and relays it to the external management system, thus resolving the contradiction between direct device usage and failure detectability.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Productivity

If a virtual machine directly uses a physical device, then throughput performance is improved, but the virtualization device cannot monitor the physical device status

Engineering Contradiction:
Improvethroughput performanceVSAvoiddevice status monitoring
Core Design Contradiction:
ProductivityVSDifficulty of detecting and measuring

Solution Approach 1:

The patent replaces the traditional mechanical/direct monitoring approach with a software-based information relay mechanism. Instead of the virtualization device directly monitoring the physical device through hardware interfaces, the system uses a software agency within the virtual machine to collect and transmit status information. This substitution allows the virtual machine to maintain direct control for high throughput while enabling indirect monitoring capabilities for the virtualization manager.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

3Speed

If PCI passthrough function is used to make virtual machine directly use physical network interface, then communication speed is improved, but only virtual machine manager can detect failure not virtualization device manager

Engineering Contradiction:
Improvecommunication speedVSAvoidfailure information
Core Design Contradiction:
SpeedVSLoss of information

Solution Approach 1:

The patent segments the failure detection function into a separate, dedicated component (failure detection agency) that operates independently within the virtual machine environment. This segmentation allows the physical device to be directly assigned to the virtual machine for high-speed communication while the segmented detection agency specifically handles monitoring tasks. The agency collects failure information and transmits it externally, preventing information loss and enabling the virtualization device manager to receive failure notifications.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP2975524B1Information processing device
Publication Date: 2019.06.05 NEC CORP
  • EP2975524B1 patent drawingFigure 1~2
  • EP2975524B1 patent drawingFigure 3~4
  • EP2975524B1 patent drawingFigure 5

AI summary

An information processing device includes: a virtual machine built in the information processing device and able to use a physical device included by the information processing device; and an information processing device failure managing means for detecting a failure in the information processing device. The virtual machine includes: a virtual machine failure managing means for detecting a failure in the physical device which the virtual machine can use; and a failure notifying means for notifying occurrence of a failure in the physical device detected by the virtual machine failure managing means to the information processing device failure managing means.