Container Group Storage Migration with Checkpoint Notification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional storage services experience prolonged interruptions and data unavailability during container group rescheduling and upgrading due to forced dumping processes and incorrect readiness reporting by container service choreography tools, leading to inefficient storage performance.
Innovation Solution
A method that involves stopping cache data dumping at the source container group, sending a notification with a checkpoint to the target container group, and updating ownership to the target container group, ensuring seamless migration and reducing storage service interruptions by maintaining the source container group in a service state until the target is ready.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If container group rescheduling and upgrading is performed using traditional forced dumping processes, then container group migration can be completed, but storage service interruption time increases and storage performance deteriorates
Solution Approach 1:
The system performs preliminary actions by stopping the dumping process before actual migration begins, preparing the memory table in advance. The source container group maintains its service state while the target container group is being set up, so that when migration occurs, the data is already prepared and can be transferred immediately without service interruption.
Solution Approach 2:
The patent introduces an intermediary mechanism where the source container group continues to serve as the active service provider during the transition period. The target container group is prepared in advance but does not take over until the source is ready to migrate, ensuring continuous service availability throughout the migration process.
2Productivity
If traditional forced dumping is used during container group rescheduling, then data can be migrated to the target container group, but storage performance decreases due to service unavailability
Solution Approach 1:
The source container group maintains continuous service operation throughout the migration process. By stopping the dumping process in advance and keeping the source container group in service state, the system ensures that storage services continue uninterrupted, thereby maintaining both high productivity and reliability during the migration.
3Extent of automation
If container service choreography tool manages container group rescheduling, then automated migration can be achieved, but readiness detection accuracy decreases leading to premature service switching
Solution Approach 1:
The system implements a feedback mechanism where the container service choreography tool receives explicit readiness signals from the source container group before initiating migration. The source container group actively communicates its service state and migration readiness, providing accurate feedback that prevents premature switching and ensures the target container group is truly ready to take over.
Data Source
AI summary
Storage processes are described. For instance, dumping of cache data in a memory table at a source directory table partition in a source container group is stopped in response to receiving a close signal. A notification for creating a target directory table partition is sent to a target container group, the notification including a checkpoint at the source directory table partition. The cache data in the memory table is sent to the target container group, and the source container group is closed in response to a determination that an ownership of the target directory table partition is updated to the target container group. A rescheduling and/or upgrading solution is thus proposed for a source container group, so that the source container group is still in a service state during the rescheduling and/or upgrading solution, which improves the performance of storage services and reduces the data unavailable time.