Storage Checkpointing for Independent Applications
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current enterprise and wireless systems lack mechanisms for protecting multi-application services, where individual applications from different vendors are loaded at different times, and traditional storage checkpointing technologies affect unrelated applications on a server.
Innovation Solution
A method and system for storage checkpointing that allows individual applications to restore storage without affecting other applications, using a coordinator to manage registration, fault detection, and recovery across independent applications, with support for both fork() and exec() calls, and transparent migration of application groups.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional storage checkpointing is implemented at the storage device level, then storage protection is achieved, but all applications on the server are affected including unrelated applications
Solution Approach 1:
The patent segments the storage checkpointing process by application, creating separate checkpoint records for each application that utilizes storage. The storage management module maintains a mapping between applications and their specific storage devices, allowing selective checkpointing and restoration of individual applications without affecting others. This resolves the contradiction by dividing the monolithic storage protection approach into application-specific segments.
Solution Approach 2:
The patent implements local quality by providing customized checkpointing characteristics for each application. The system allows different checkpointing policies, frequencies, and storage locations for different applications based on their specific requirements. Each application can have its own checkpoint record with application-specific parameters, enabling tailored protection strategies that don't impact other applications.
2Reliability
If storage checkpointing is implemented for multi-application services, then service availability is improved, but device complexity increases due to coordinating multiple independent applications
Solution Approach 1:
The patent merges the checkpointing functionality into a unified storage management module that coordinates multiple applications centrally. This module maintains a database of checkpoint records for all applications and provides a single point of control for checkpointing operations. By combining scattered checkpointing logic into one centralized manager, the system reduces overall complexity while improving service availability through coordinated recovery.
Solution Approach 2:
The storage management module is designed with universal functionality to handle checkpointing for multiple different applications simultaneously. It provides a standardized interface and common checkpointing mechanism that works across diverse applications, eliminating the need for application-specific checkpointing implementations and reducing system complexity.
3Reliability
If checkpointing is performed for the entire storage device, then comprehensive protection is achieved, but restoration time increases when only one application needs recovery
Solution Approach 1:
The patent segments storage protection into application-specific checkpoint records, allowing selective restoration of individual applications. When an application fails, only its specific checkpoint record is restored along with the minimal required storage data, rather than restoring the entire storage device. This segmentation dramatically reduces restoration time while maintaining comprehensive protection through individualized checkpointing.
Solution Approach 2:
The system performs partial checkpointing by creating checkpoint records only for applications that actually utilize storage resources. During restoration, only the necessary portion of storage data corresponding to the failed application is restored, avoiding the excessive action of restoring entire storage devices or unrelated application data.
Data Source
AI summary
A method and system for storage checkpointing of an independent computer application. The independent computer application is launched by a coordinator; and the coordinator installs at least one of an exec interceptor and a fork interceptor. The coordinator also installs at least one file operations interceptor for all file operations and registers the independent computer application with the coordinator. The independent computer application is run and the at least one file operations interceptor is called upon encountering a file operation. The file operations interceptor logs a file event in a file operations database and passes the operation to at least one of a file system, an operating system, at least one or more device drivers, and a storage disk via a storage interface. The file operations interceptor also verifies that the file operation has been issued.