Active-Passive Database Backup With Dynamic Stream Allocation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The use of limited computing resources for data protection operations on databases can impact the processing of other tasks, leading to inefficiencies in resource utilization.
Innovation Solution
A data protection system that selectively allocates backup streams to computing nodes in a database cluster, leveraging load balancing capabilities and performing intelligent stream distribution, while ensuring efficient resource allocation and handling node availability during backups.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data protection operations are performed on databases using computing resources, then data protection is achieved, but other processing tasks are limited due to resource consumption
Solution Approach 1:
The patent segments database clusters into active and passive nodes, and further divides backup operations into multiple streams that can be distributed across different nodes. This segmentation allows data protection operations to be performed without blocking all other processing tasks, as backup streams can be allocated to passive nodes or scheduled during low-activity periods, thereby resolving the contradiction between data protection reliability and overall system productivity.
Solution Approach 2:
The patent implements dynamic resource allocation where backup streams can be dynamically assigned to different computing nodes based on real-time cluster state, node availability, and workload conditions. This dynamic approach allows the system to adapt backup operations to current system conditions, ensuring data protection while minimizing impact on other processing tasks through flexible resource utilization.
2Reliability
If backup streams are allocated to computing nodes, then data protection operations can be performed, but node availability and cluster state changes complicate the process
Solution Approach 1:
The patent implements feedback mechanisms where the data protection system continuously monitors cluster state changes, node availability, and backup stream status. When nodes are added, removed, or fail over, the system receives feedback and automatically adjusts backup stream allocations accordingly. This feedback loop ensures data protection operations remain reliable despite changing node availability conditions.
Solution Approach 2:
The patent performs preliminary discovery of database clusters and their configurations before initiating backup operations. The system proactively identifies active and passive nodes, understands cluster topology, and pre-configures backup stream allocations based on anticipated node availability. This preliminary action prepares the system to handle node changes more effectively during actual backup operations.
Data Source
AI summary
A method for managing data protection operations includes obtaining, by a data protection system, a policy request for configuring a data protection policy on the database system, wherein the database system comprises at least one active computing node, in response to obtaining the policy request: making a first determination that a first computing node of the database system has been discovered via deep discovery, updating a selection of computing nodes with the first computing node, making a second determination that a second computing node of the database system has not been discovered via the deep discovery, not including the second computing node in the selection, and performing a protection configuration on the selection of computing nodes to install the data protection policy to each computing node in the selection of computing nodes, and wherein the data protection policy comprises storage location information used for backing up a database.