Live VM Update via Priority-Based Resource Reallocation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing live update technologies fail in resource-constrained systems, leading to downtime or recovery requirements for business-critical workloads due to insufficient system resources.

Innovation Solution

A method that reallocates resources from lower-priority virtual machines to create a new VM with sufficient resources for live updates, allowing non-disruptive migration of workloads to the new VM, and returning resources after the update.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If live update is performed on a running system, then software updates can be applied without downtime, but the system requires sufficient available resources which may not be present in resource-constrained systems

Engineering Contradiction:

Improvecontinuous operation of business-critical workloadsVSAvoidavailable system resources

Core Design Contradiction:

ReliabilityVSQuantity of substance

Solution Approach 1:

The patent introduces an intermediary VM (virtual machine) as a mediator between the original VM and the updated state. This intermediary VM is created with the updated software version while the original VM continues running, allowing the update to be applied without interrupting business-critical workloads. The intermediary acts as a bridge that enables the transition from old to new software state.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The update process is segmented into distinct phases: creating an intermediary VM with updated software, migrating workloads gradually, and maintaining the original VM running until the update is complete. This segmentation allows resource allocation to be managed in stages, with resources being reallocated from the original VM to the intermediary VM as needed, rather than requiring all resources upfront.

Inventive Principle:

Principle #1Segmentation

2Quantity of substance

If resources are reallocated from existing VMs to create a new VM for live update, then sufficient resources become available for the update, but the donor VMs experience reduced resource availability

Engineering Contradiction:

Improveresources for new VMVSAvoidoperation of donor VMs

Core Design Contradiction:

Quantity of substanceVSReliability

Solution Approach 1:

The patent implements dynamic resource allocation where resources are temporarily borrowed from donor VMs during the update process and then returned after completion. The resource allocation is dynamic rather than static, allowing the system to flexibly adjust resource distribution based on the update lifecycle stage. This ensures donor VMs maintain their resource levels during and after the update.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

Resources are temporarily discarded from donor VMs during the creation and initialization phase of the new VM, but are systematically recovered and returned to donor VMs once the new VM is successfully created and workloads are migrated. This temporary discarding and subsequent recovery allows the update to proceed without permanently impacting donor VM operations.

Inventive Principle:

Principle #34Discarding and recovering

3Reliability

If the live update process fails due to insufficient resources, then recovery is required which causes downtime, but preventing the update avoids software security and functionality improvements

Engineering Contradiction:

Improvesystem availabilityVSAvoidsoftware update capability

Core Design Contradiction:

ReliabilityVSAdaptability or versatility

Solution Approach 1:

The patent prepares cushioning measures in advance by creating the intermediary VM with all necessary resources allocated before the update process begins. This prior preparation ensures that if resources become constrained during the update, the intermediary VM already has the buffer needed to complete the update successfully, preventing failures that would require recovery and cause downtime.

Inventive Principle:

Principle #11Beforehand cushioning (Prior cushioning)

Data Source

PatentUS20250377885A1Operating system live update in a resource constrained system

Publication Date: 2025.12.11 INTERNATIONAL BUSINESS MACHINE CORPORATION

AI summary

A method, computer system, and computer program product are provided to perform operating system live update in a resource constrained computer system. A computer system orchestrator verifies that contents of a profile definition of each virtual machine (VM) includes a valid assigned priority. Orchestrator verifies sufficient free resources are available in the computer system to create a new VM corresponding to an original VM. Based on there not being sufficient free resources, beginning with a donor VM having a lowest priority and continuing to the donor VM having a next higher priority, moving resources from the donor VM to the new VM until the new VM has resources corresponding to the original VM. Updating software on the original VM. Exporting the root volume group to a new VM that is activated using the updated root volume group. The workload is non-disruptively migrated from the original VM to the new VM.