Hot Standby Update via Backup System Intermediary
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems face challenges in maintaining a hot standby for continuous data protection-enabled objects, particularly in achieving near-zero Recovery Point Objective (RPO) and Recovery Time Objective (RTO) due to resource constraints and communication issues between backup and remote data recovery systems, which can lead to delays and inconsistencies in data updates.
Innovation Solution
The implementation of a write filter that intercepts inputs/outputs, provides a stream of source system storage changes to a backup system, and uses snapshot markers to generate reference restoration points, allowing the remote data recovery system to independently update the hot standby, thereby reducing the influence of resource constraints and communication issues.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data is directly streamed from the source system to a hot standby, then near-zero RPO and RTO can be achieved, but the performance of the source system deteriorates because resources are consumed for streaming
Solution Approach 1:
A backup system is introduced as an intermediary between the source system and remote data recovery system. The backup system receives storage changes from the source system and forwards them to the remote data recovery system, allowing the source system to focus on its primary operations while the backup system handles the data protection traffic.
Solution Approach 2:
The data protection functionality is segmented into separate systems: the source system generates changes, the backup system manages the stream, and the remote data recovery system maintains the hot standby. This segmentation prevents the source system from being burdened with hot standby maintenance tasks.
2Reliability
If resources are allocated to stream data to hot standby, then near-zero RPO and RTO are achieved, but resource availability for source system operations decreases
Solution Approach 1:
The backup system acts as a mediator that consumes minimal resources from the source system while maintaining the ability to propagate changes to the hot standby. This intermediary approach allows data protection without significantly impacting source system resource availability.
3Speed
If the remote data recovery system updates the hot standby independently using snapshot markers, then communication issues and delays are reduced, but system complexity increases
Solution Approach 1:
Snapshot markers are inserted into the change stream in advance at the source system or backup system. These markers pre-identify recovery points, allowing the remote data recovery system to independently process and apply changes without waiting for coordination messages, thereby speeding up hot standby updates.
4Loss of time
If parallel ingestion and replication of storage changes is implemented, then hot standby update time is reduced, but communication requirements increase
Solution Approach 1:
The parallel processing is segmented into ingestion operations and replication operations that can occur simultaneously. The backup system ingests changes from the source system while the remote data recovery system replicates them to the hot standby, reducing total update time through concurrent execution of these distinct tasks.
Data Source
AI summary
A remote data recovery system is determined to be unsuitable for communications. A stream of source system storage changes associated with an object is received at a backup system from a source system while the remote data recovery system is unsuitable for communications. The backup system is utilized to generate one or more reference restoration points based on the stream of source system storage changes associated with the object. The remote data recovery system is determined to be suitable for communications. In response to determining that the remote data recovery system is suitable for communications, a hot standby of the object hosted by the remote data recovery system is updated to a reference restoration point generated by the backup system prior to the remote data recovery system becoming suitable for communications.