On-demand Virtual Node Provisioning for Server Cluster Failover

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing server cluster failover solutions consume resources and incur additional costs due to passive servers waiting for failures, which can be inefficient and power-intensive.

Innovation Solution

Implementing a server cluster architecture where virtual nodes are only provisioned on-demand, with physical nodes managing virtual machines to seamlessly fail over applications to active virtual nodes upon failure, minimizing resource usage and power consumption.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If passive servers are deployed to take over failures from active servers, then reliability is improved, but resource consumption and power usage increase

Engineering Contradiction:

Improvefailure recoveryVSAvoidpower consumption

Core Design Contradiction:

ReliabilityVSUse of energy by moving object

Solution Approach 1:

The patent implements dynamic provisioning of virtual nodes based on failure events. Instead of having static passive servers continuously running, the system dynamically activates virtual nodes only when failures occur, allowing the system to adapt its resource consumption to actual needs while maintaining reliability

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system changes the operational state of virtual nodes from continuously active to on-demand active. By changing the parameter of virtual node provisioning from static to dynamic, the system reduces power consumption while maintaining the ability to recover from failures when needed

Inventive Principle:

Principle #35Parameter changes

2Reliability

If passive servers continuously consume resources waiting for failures, then failover capability is improved, but productivity decreases

Engineering Contradiction:

Improvefailover capabilityVSAvoidresource allocation efficiency

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The system dynamically provisions virtual nodes based on actual failure events rather than maintaining static standby capacity. This dynamic approach improves productivity by allocating resources only when needed for failover, eliminating the waste of continuously consuming resources in advance

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The patent creates virtual copies of server functionality on-demand rather than maintaining physical copies. By using virtualization to create software-based copies only when failures occur, the system maintains failover capability while dramatically improving resource allocation efficiency

Inventive Principle:

Principle #26Copying

3Loss of energy

If virtual nodes are provisioned on-demand, then resource consumption is reduced, but failover time may increase

Engineering Contradiction:

Improveresource consumptionVSAvoidfailover time

Core Design Contradiction:

Loss of energyVSLoss of time

Solution Approach 1:

The system performs preliminary configuration of virtual nodes in advance, preparing the infrastructure and templates needed for failover. When failures occur, pre-configured virtualization capabilities enable rapid provisioning, reducing actual failover time while maintaining resource efficiency

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent changes the provisioning parameter from continuous availability to event-triggered availability. By using virtualization technology, the system can provision nodes rapidly in response to failures, balancing resource consumption reduction with acceptable failover time through efficient virtual machine deployment

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS7814364B2On-demand provisioning of computer resources in physical/virtual cluster environments

Publication Date: 2010.10.12 DELL PROD LP

AI summary

A server cluster comprises a physical node and a virtual host node. The physical node includes an active node running an application. The virtual host node includes an inactive virtual node. The virtual node is activated upon failure of the active node and the application is failed over to the virtual node.