Node Migration for Shared Infrastructure Service Continuity

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

In information handling systems where multiple nodes share physical components, servicing one of these components often requires powering down multiple nodes, leading to disruptions in service as applications experience downtime.

Innovation Solution

Implementing a method where the system state and system image of nodes on a system board requiring service are transferred to a target processor on another system board before the originating system board is serviced, thereby allowing nodes to operate uninterrupted.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Volume of moving object

If physical components are shared among multiple nodes to increase system density, then space utilization is improved, but service disruption occurs when servicing shared components

Engineering Contradiction:

Improvesystem densityVSAvoidservice continuity

Core Design Contradiction:

Volume of moving objectVSReliability

Solution Approach 1:

The system performs preliminary actions by identifying a target processor before the originating system board is removed, preparing the target processor to receive and execute the node's system image, thereby ensuring service continuity without interruption

Inventive Principle:

Principle #10Preliminary action

2Reliability

If system board servicing is performed on shared infrastructure nodes, then component reliability is improved, but application downtime increases

Engineering Contradiction:

Improvecomponent reliabilityVSAvoidapplication downtime

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system maintains continuity of useful action by transferring the node's operational state to a target processor, allowing the application to continue executing without interruption while the original system board is serviced

Inventive Principle:

Principle #20Continuity of useful action

Solution Approach 2:

The system creates a copy of the node's system image and transfers it to a target processor, enabling the application to run on an alternative processor while the original board is being serviced

Inventive Principle:

Principle #26Copying

Data Source

PatentUS9354993B2System and method to reduce service disruption in a shared infrastructure node environment

Publication Date: 2016.05.31 DELL PROD LP

AI summary

A method of reducing downtime in a node environment is disclosed. The method includes identifying an originating system board of a plurality of system boards that requires service where the originating system board includes a node operating on a processor. The method further includes identifying a target system board of the plurality of system boards where the target system board includes a target processor. The method further includes transferring operation of the node to the target processor before the originating system board is serviced, and operating the node on the target processor.