Cloud Management Platform Restoration via Journaling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Sophisticated cloud management platforms (CMPs) face challenges in consistently restoring to a previous state after failures, often resulting in data loss or undesirable outcomes due to the use of diverse and unrelated components with different backup solutions, which complicates achieving a consistent state across heterogeneous systems.
Innovation Solution
A Consistent Restoration System (CRS) utilizing a journaling algorithm to track data modifications across components, allowing for the selection and automatic or semi-automatic restoration to a determined optimal point, leveraging a Representational State Transfer (REST) API to identify and log data modifications, and using an independent restoration system to restore the CMP to a consistent state.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If diverse and unrelated components with different backup solutions are used in cloud management platforms, then the platform can integrate multiple specialized components for specific functionality, but achieving a consistent state across heterogeneous systems becomes complicated and results in data loss or undesirable outcomes during restoration
Solution Approach 1:
The patent introduces a restoration management component as an intermediary that coordinates restoration operations across diverse components. This mediator maintains a restoration plan that tracks the state of each component and ensures they are restored in the correct sequence and to consistent states, resolving the contradiction between component diversity and restoration consistency.
Solution Approach 2:
The patent implements preliminary actions by creating and maintaining a restoration plan before failures occur. The restoration plan pre-defines relationships between components, dependency structures, and restoration sequences, enabling consistent restoration across heterogeneous systems without ad-hoc coordination during crisis situations.
2Loss of information
If all components are restored to ensure system consistency, then data loss is minimized, but restoration time and resource consumption increase significantly
Solution Approach 1:
The patent applies partial action by restoring only the necessary components identified in the restoration plan rather than all components. The restoration management component analyzes the failure impact and coordinates restoration of minimal necessary components to achieve system consistency, reducing restoration time while preventing data loss through selective restoration based on component dependencies.
3Reliability
If a centralized restoration system is implemented to coordinate all components, then consistent restoration is achieved, but system complexity and coordination overhead increase
Solution Approach 1:
The patent segments the restoration system into autonomous components, each capable of self-restoration based on instructions from the restoration management component. The restoration plan divides the system into manageable units with defined relationships, allowing coordinated restoration without requiring a fully centralized control mechanism for every operation.
Data Source
AI summary
In some examples, a method includes identifying, via the use of a Representational State Transfer (REST) Application Programming Interface (API) call, a modification of persistent data for a Cloud Management Platform (CMP); storing, in a persistent log for the CMP, information about the data modification, including an operation that modified the data, a component of the CMP that modified the data, and the time of modification; determining, as a result of a failure of the CMP, a restoration point for the CMP based on the persistent log; and restoring the CMP to the determined restoration point using an independent restoration system of the component that modified the data.