Virtual Machine Mirror File Pulling Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing cloud computing platform architecture faces high loads and slow virtual machine creation speeds due to the heavy loads on servers when pulling large mirror files for virtual machine creation, particularly with Windows operating systems, where the mirror files are around 10G in size.
Innovation Solution
A method and device that obtain a mirror pulling instruction with a timestamp, determine if only a base mirror file is needed, and if so, retrieve it locally; otherwise, retrieve both the base and incremental files matching the timestamp, optimizing data retrieval by reducing the amount of data to be pulled and distributing the load between local storage and the mirror server.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the virtual machine management module retrieves the mirror file through the virtual machine mirror management module and pulls the mirror file from the mirror server, then the virtual machine can be created, but the loads on the server and mirror server become very heavy and the creation speed becomes slow
Solution Approach 1:
The patent applies preliminary action by pre-distributing base mirror files to multiple physical machines before they are needed for virtual machine creation. When a virtual machine needs to be created, the system can quickly retrieve the base mirror file from local physical machines rather than pulling large files from the mirror server, significantly reducing server load and creation time.
Solution Approach 2:
The patent segments the mirror file into two parts: base mirror files (common to multiple virtual machines) and incremental files (specific to individual virtual machines). This segmentation allows the system to store only the base portion locally on physical machines while keeping incremental updates on the mirror server, reducing the amount of data that needs to be transferred and stored locally.
2Productivity
If the mirror file is pulled from the mirror server for each virtual machine creation, then the virtual machine can be created, but the retrieval period becomes long due to large file size
Solution Approach 1:
The system performs preliminary distribution of base mirror files to physical machines in advance, so that when virtual machine creation is needed, the base file is already available locally, eliminating the need to retrieve large files from the mirror server during the creation process.
Solution Approach 2:
The patent extracts the base mirror file (which is common and large) from the complete mirror file package and stores it separately on physical machines. Only the smaller incremental files need to be retrieved from the mirror server, significantly reducing retrieval time.
3Speed
If the base mirror file is stored locally on physical machines, then the retrieval speed improves, but the storage space requirement on physical machines increases
Solution Approach 1:
The patent segments the mirror file into base mirror files (stored locally on physical machines) and incremental files (stored on the mirror server). This segmentation allows the system to store only the essential base portion locally, reducing the storage burden compared to storing complete mirror files, while still achieving fast retrieval speeds.
Solution Approach 2:
The patent applies local quality by storing different types of data in different locations: base mirror files are stored locally on physical machines for fast access, while incremental updates are kept on the mirror server. This optimized storage strategy balances retrieval speed with storage space requirements.
Data Source
AI summary
A method and device for pulling a virtual machine mirror file includes: obtaining a mirror pulling instruction comprising a timestamp of a mirror file; determining whether the mirror pulling instruction instructs only a base mirror file; if yes, obtaining locally the base mirror file previously dispatched to physical machines; and if not, obtaining locally the base mirror file previously dispatched to physical machines, and obtaining an incremental file matching the timestamp of the mirror file. The amount of data of the virtual machine mirror file to be pulled by the virtual machine mirror management module is reduced by using the base mirror file previously dispatched to physical machines, such that the pulling period is shortened, and the loads of a server where the virtual machine mirror management module is located and the loads of the mirror server are decreased. Accordingly, the creation of the virtual machine may be accelerated.