Persistent Data Management in Elastic Virtual Machine Instances
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current cloud computing solutions for IaaS platforms, such as Amazon EC2 and IBM ISAAC, face challenges in managing persistent data for pre-installed applications in elastic virtual machine instances, leading to data loss upon instance termination and inefficient data management processes.
Innovation Solution
A method and apparatus that involve obtaining a persistent data point for each function component of an application, creating a storage volume, attaching it to the virtual machine instance, and mounting it to a file directory, with features like initial data backup and restoration, and recording mapping relationships to ensure data persistence across instance launches and terminations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If applications are pre-installed in the master virtual machine image and data is stored locally using Copy-On-Write technique, then maintenance costs and disk data are reduced, but data persistence is lost when the instance is terminated or hardware fails
Solution Approach 1:
The patent segments the storage system into two parts: the master virtual machine image for read-only shared data and applications, and separate storage volumes for persistent data. This segmentation allows the master image to be efficiently shared while ensuring persistent data is stored separately in the storage domain, resolving the contradiction between maintenance efficiency and data persistence.
Solution Approach 2:
The patent introduces storage volumes as an intermediary between the virtual machine instances and the storage domain. These volumes act as a mediator that enables persistent data storage while maintaining the Copy-On-Write efficiency of the master image, allowing data to be written persistently without affecting the shared master image.
2Reliability
If storage volumes are created and attached after virtual machine instance launch, then persistent storage is enabled, but pre-installed applications cannot store their data persistently
Solution Approach 1:
The patent performs preliminary actions by pre-configuring persistent data points in the master virtual machine image and pre-creating storage volumes before launching virtual machine instances. This allows pre-installed applications to automatically have their data stored persistently from the start, eliminating the need for manual post-deployment configuration while maintaining deployment efficiency.
Solution Approach 2:
The patent implements a feedback mechanism where the system automatically detects pre-installed applications, identifies their persistent data requirements, and configures appropriate storage volumes. This automated feedback loop ensures that persistent storage is properly configured for pre-installed applications without requiring manual intervention, resolving the contradiction between data persistence and deployment efficiency.
3Reliability
If manual backup solutions are implemented for pre-installed application data, then data persistence is achieved, but the process becomes complex and requires additional management overhead
Solution Approach 1:
The patent enables self-service by automatically configuring persistent storage for pre-installed applications based on pre-defined persistent data points in the master image. The system automatically creates storage volumes, attaches them to instances, and mounts them to the correct directories without requiring manual backup configuration, thereby achieving data persistence while minimizing management complexity.
Data Source
AI summary
A method and apparatus for managing the persistent data of a pre-installed application in an elastic virtual machine instance is disclosed, the method comprising: in response to installing an application into a master virtual machine image, obtaining a persistent data point of a function component of the application, the persistent data point comprising a file directory for storing the persistent data of the function component; in response to launching an elastic virtual machine instance from the master virtual machine, creating a storage volume, attaching the storage volume to the elastic virtual machine instance, and mounting the storage volume to the file directory based on the persistent data point. Other features and aspects may be realized, depending upon the particular application.