Virtual Machine Snapshot Synchronization for Offline Work
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In cloud computing environments, uploading large files to a data center is slow due to slower upload speeds compared to download speeds, and existing methods require uploading the entire virtual machine (VM) or maintaining multiple copies, which is inefficient in terms of bandwidth and storage.
Innovation Solution
The system allows modifications to virtual machines by taking snapshots, computing differences between the original and modified VMs, and uploading only these changes to the data center, enabling efficient storage and transmission without requiring the entire VM to be uploaded or downloaded.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the entire virtual machine is uploaded to the data center, then the VM can be properly installed and configured, but the upload time becomes excessively long due to slow upload speeds
Solution Approach 1:
The patent extracts only the necessary components for VM installation and configuration from the complete VM image. Instead of uploading the entire VM, the system identifies and uploads only the essential installation files and configuration data, significantly reducing upload time while maintaining installation completeness
Solution Approach 2:
The VM installation process is segmented into multiple phases: uploading installation files first, then configuring the VM locally, and finally syncing only the necessary configuration data to the data center. This segmentation allows the upload phase to be much shorter while still achieving complete VM setup
2Ease of operation
If multiple copies of the virtual machine are maintained for offline work, then offline access is enabled, but storage requirements increase significantly
Solution Approach 1:
The patent implements a nested storage structure where the local VM copy is embedded within the cloud storage system. The local copy stores only differential changes and configuration data, while the complete VM image remains stored in the cloud. This nested approach enables offline work with minimal local storage requirements
Solution Approach 2:
The system performs preliminary configuration of the VM locally using cached installation files and configuration templates before syncing to the data center. This preliminary action enables offline work capability without requiring complete VM copies to be stored locally, reducing storage requirements
3Reliability
If the virtual machine is modified locally and the entire VM is re-uploaded, then the modifications are saved to the data center, but bandwidth usage increases unnecessarily
Solution Approach 1:
The system extracts only the modified portions of the VM from the local copy and uploads only these differential changes to the data center. By identifying and extracting only the changed files and configuration data, bandwidth usage is significantly reduced while ensuring all modifications are persisted
Solution Approach 2:
The patent implements a feedback mechanism that compares the local VM copy with the cloud-stored VM to identify only the modified sections. This feedback loop ensures that only necessary changes are uploaded, optimizing bandwidth usage while maintaining modification persistence
Data Source
AI summary
Virtualization snapshots are utilized for efficiently synchronizing virtual machines present on a client system and a server system. A snapshot of a virtual machine (VM) is taken by the client system before making modifications to the VM. The modifications to the VM are stored by the virtualization snapshot separate from the VM itself. The modified portions of the VM are communicated by the client system to the server system. The server system applies the modifications to a locally stored VM that corresponds to the original VM on the client system. The modified VM on the server system matches the modified VM on the client. As a result synchronization between the VMs on client system and the server system is obtained without transferring the entire VM multiple times. The server system performs format conversions between a format of the client system and the server system on the VM.