Phased Product Migration Using Intermediate Instances and Rollback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for migrating computing product versions require a big-bang approach, leading to lengthy outages, high failure rates, and disruptions in continuous integration and deployment pipelines, with no rollback option.
Innovation Solution
A method and system for automatically migrating projects in phases using an AI model to identify projects, create intermediate instances, and perform data file transfers between instances, with metric comparison and notification of successful migration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a big-bang migration approach is used to migrate the entire database at once, then the migration can be completed in a single operation, but the outage time increases to 48 hours and the failure rate becomes high
Solution Approach 1:
The patent divides the migration process into multiple phases, migrating a subset of projects first to a first intermediate instance, then migrating those same projects to a second intermediate instance. This segmented approach reduces the risk of complete migration failure while maintaining overall migration progress, directly addressing the high failure rate issue of big-bang migration.
Solution Approach 2:
The patent performs preliminary migration actions by first copying data to intermediate instances before finalizing the migration. This allows validation and rollback capabilities, reducing the risk of complete migration failure while still achieving the ultimate migration goal.
2Loss of time
If a big-bang migration approach is used, then the entire database can be migrated at once, but the outage time increases to 48 hours for each line of business
Solution Approach 1:
The patent segments the migration process into multiple phases with intermediate instances, allowing continuous operation of non-migrated projects while migrated projects are transitioned. This reduces the effective outage time from 48 hours to a fraction of that time, as only subsets of projects are migrated at each phase.
Solution Approach 2:
The patent introduces intermediate instances as intermediary storage locations between the source and final destination. These intermediates allow data to be staged and validated before final migration, reducing the impact on production systems while maintaining migration progress.
3Productivity
If a big-bang migration is performed, then the entire database is copied at once, but continuous integration and deployment pipelines are impacted and production deployments are not allowed
Solution Approach 1:
The patent divides the database into multiple project subsets that can be migrated independently to intermediate instances. This allows CI/CD pipelines to continue operating on non-migrated projects while migration occurs on specific subsets, maintaining pipeline continuity and adaptability throughout the migration process.
Solution Approach 2:
The patent performs preliminary migration to intermediate instances before final deployment, allowing validation and testing without impacting production CI/CD pipelines. This preliminary action enables migration progress while preserving pipeline adaptability.
4Reliability
If the entire database is migrated at once, then complete migration can be achieved, but rollback to initial versions is not possible
Solution Approach 1:
The patent segments the migration into phases with intermediate instances, where each phase can be independently rolled back if problems occur. This maintains migration completeness over time while enabling rollback capability at each segment, directly addressing the inability to rollback in big-bang migration.
Solution Approach 2:
The patent creates intermediate instances as cushioning layers between the source and final destination. These intermediates serve as rollback points, allowing the system to revert to previous states if migration issues arise, thereby preserving both completeness and rollback capability.
Data Source
AI summary
A method and a system for automatically migrating projects of a computing product in phases for a version upgrade are provided. The method includes: receiving at least one input relating to projects of the computing product; identifying, based on the at least one input, at least one project of the first version of the computing product to be migrated; deploying a first intermediate instance for the first version of the computing product; copying at least one first data file of the identified at least one project from a source instance to the first intermediate instance; importing the identified at least one project from the at least one first data file into the first intermediate instance; modifying the first intermediate instance to a second intermediate instance for the second version; copying at least one second data file; and importing the at least one identified project of the second version.