Replication Appliance Splitter for Consistency Group Resource Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current replication solutions, such as continuous and snapshot-based replication, consume excessive resources and generate high traffic loads, limiting their efficiency for less critical applications, where lower granularity protection is sufficient.
Innovation Solution
The method involves dynamic determination of whether to perform full consistency group replication or mini-CG replication based on predefined application criteria, using a splitter function to provide either data or metadata to replication appliances, reducing resource consumption by transferring only metadata for mini-CG replication and snapshots of changed data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If continuous replication is used for all applications, then data protection reliability is improved, but resource consumption and traffic load increase excessively
Solution Approach 1:
The patent segments applications into different consistency group types (full CG and mini-CG) based on their replication requirements. This segmentation allows the system to apply different replication strategies to different application groups, providing high reliability for critical applications while reducing resource consumption for less critical ones.
Solution Approach 2:
The system dynamically determines the appropriate replication mode for each consistency group based on predefined application criteria. The splitter function automatically routes data or metadata to replication appliances based on the determined replication mode, allowing the system to adapt resource allocation to actual application needs rather than applying a static replication strategy to all applications.
2Reliability
If full consistency group replication is performed for all applications, then data protection is improved, but device complexity and management overhead increase
Solution Approach 1:
The patent divides consistency groups into two categories: full CG for applications requiring complete data protection and mini-CG for applications where lower granularity protection is sufficient. This segmentation simplifies management by allowing different replication modes to be applied to different application groups based on their specific requirements.
Solution Approach 2:
The system changes the replication parameter (full data vs. metadata only) based on the consistency group type. For mini-CG applications, the system uses metadata-only replication which significantly reduces the amount of data processed and managed, thereby reducing device complexity and management overhead while still providing adequate protection.
3Use of energy by moving object
If metadata-only replication is used for mini-CG, then resource consumption is reduced, but replication granularity is reduced
Solution Approach 1:
The patent applies different replication granularities to different consistency groups based on their specific requirements. Full CG applications receive fine-grained replication with complete data protection, while mini-CG applications receive coarser-grained metadata-only replication. This local quality approach ensures that each application group receives the appropriate level of replication granularity for its needs.
Data Source
AI summary
Methods and apparatus are provided for consistency group replication with reduced resource consumption. An exemplary method comprises: obtaining, by one or more first replication appliances at a first location, one of (i) data generated by a monitored application; and (ii) metadata for each write operation generated by the monitored application, wherein a splitter function provides the data or metadata to the one or more first replication appliances based on a dynamic determination of whether the monitored application requires replication based on the data or the metadata based on predefined application consistency group type criteria; obtaining, by one or more of the first replication appliances, a snapshot comprising the data associated with each of the write operations since a prior snapshot based on the one of the data and the corresponding metadata; and providing the snapshot to one or more second replication appliances at a different location for storage in a replication journal.