Hypervisor Virtual-Memory File System Update Handoff
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Hypervisor updates often disrupt virtual-machine operations due to the need for rebooting and data copying, which can result in significant network bandwidth consumption and performance penalties, making them undesirable and sometimes unacceptable.
Innovation Solution
The implementation of a virtual-memory file system (VMemFS) allows for the transfer of data pointers between hypervisor versions on the same machine, minimizing the need for data copying and enabling efficient hand-off of virtual machines during updates, thus reducing operational interruptions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If hypervisor updates are performed by copying data and rebooting, then the hypervisor can be updated to newer versions, but virtual-machine operations are interrupted and network bandwidth is consumed
Solution Approach 1:
The patent uses copying to create a copy-on-write layer that references the original disk image. This allows the hypervisor to be updated by copying metadata pointers rather than the entire disk image, enabling version updates without interrupting virtual machine operations or consuming excessive network bandwidth
Solution Approach 2:
The patent performs preliminary actions by pre-establishing the copy-on-write layer and preparing the updated hypervisor image before the actual update is needed. This allows the update to be applied instantly when required, avoiding interruptions to virtual machine operations
2Adaptability or versatility
If hypervisor updates are performed by checkpointing and restarting applications, then the hypervisor can be updated, but application performance is degraded and operations are interrupted
Solution Approach 1:
The patent copies only the necessary metadata and pointers rather than checkpointing entire application states. This selective copying allows the hypervisor to be updated without requiring application checkpoints or restarts, eliminating the time loss and performance degradation associated with traditional update methods
3Adaptability or versatility
If data is copied during hypervisor updates, then the hypervisor can be updated, but network bandwidth is consumed and performance is penalized
Solution Approach 1:
The patent implements a copy-on-write file system layer that copies only metadata pointers and references rather than the actual disk image data. This dramatically reduces network bandwidth consumption during hypervisor updates while still enabling version updates, as only the thin metadata layer needs to be transferred rather than the entire virtual machine disk image
Solution Approach 2:
The patent segments the disk image into the original data layer and a separate copy-on-write metadata layer. This segmentation allows the hypervisor update to affect only the metadata layer, preventing the need to copy or transfer the large bulk data and thus reducing network bandwidth consumption
Data Source
AI summary
A first hypervisor uses a first version of a virtual-memory file system (VMemFS) suspends virtual machines. A second hypervisor uses a instance of the VMemFS, the version of which may be the same or different from the first version. The VMemFS is designed so that an instance of the same or a later version of the VMemFS can read and ingest information in memory written to memory by another instance of the VMemFS. Accordingly, the second hypervisor resumes the virtual machines, effecting an update or other swap of hypervisors with minimal interruption. In other examples, the swapped hypervisors support process containers or simply support virtual memory.