Server-Side Virtual Machine Compute Reconfiguration Without Reboot Downtime
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing virtual machine provisioning processes are time-consuming, particularly in creating and configuring new VMs, which can take several minutes, and reconfiguring existing VMs often requires rebooting, leading to downtime.
Innovation Solution
A method to reconfigure virtual machines without rebooting by modifying and publishing user-specific compute artifacts, storage artifacts, and networking artifacts, allowing partially configured VMs to be quickly adapted to customer-specific settings.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If virtual machines are created and configured from scratch, then customer-specific settings can be applied, but the process takes several minutes causing long spin-up time
Solution Approach 1:
The system pre-configures virtual machines with generic settings and templates before customer-specific configuration is needed. Compute artifacts, storage artifacts, and networking artifacts are prepared in advance and can be quickly associated with customer-specific settings without requiring full reconfiguration from scratch, thus reducing spin-up time while maintaining customization capability.
Solution Approach 2:
The configuration process is divided into separate artifacts: compute artifacts, storage artifacts, and networking artifacts. These can be independently configured, modified, and associated with virtual machines. This segmentation allows partial reconfiguration without requiring complete system rebuild, enabling faster provisioning and reduced spin-up time while maintaining full customer-specific customization.
2Adaptability or versatility
If existing virtual machines are reconfigured, then customer-specific settings can be updated, but rebooting is required causing downtime
Solution Approach 1:
Configuration changes are prepared and validated in advance before being applied to running virtual machines. The system can apply updated compute artifacts, storage artifacts, and networking artifacts without requiring VM reboot, thus enabling reconfiguration while maintaining continuous operation and eliminating downtime.
Solution Approach 2:
The system enables dynamic reconfiguration of virtual machines during operation. Configuration artifacts can be modified and applied in real-time without forcing a reboot cycle. This dynamic approach allows the VM to adapt to new settings while continuing to run, eliminating the traditional reboot requirement and associated downtime.
3Adaptability or versatility
If compute artifacts are modified, then user-specific settings can be applied, but the complexity of managing multiple artifacts increases
Solution Approach 1:
The system uses a unified artifact management framework where compute artifacts, storage artifacts, and networking artifacts follow the same creation, modification, and association patterns. This universal approach simplifies management by providing consistent interfaces and workflows for all artifact types, reducing the complexity burden despite having multiple artifacts to manage.
Solution Approach 2:
The system provides feedback mechanisms that automatically track which artifacts are associated with which virtual machines and users. This feedback loop simplifies artifact management by providing clear visibility into the configuration state, enabling automated validation and reducing manual management complexity while maintaining flexible user-specific customization capabilities.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The disclosed technology is generally directed to virtual machines. A first virtual machine executing on a first virtual machine host is reconfigured without rebooting the first virtual machine. The reconfiguring includes controlling the following actions. At least one compute artifact of a plurality of compute artifacts is associated with a first user that is associated with the first virtual machine. At least one compute artifact of the plurality of compute artifacts is modified. A configuration is published to the first virtual machine host.