Pluggable Database Provisioning via Rules Engine Automation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for cloning and relocating pluggable databases in multitenant database systems require significant downtime and are costly, as they often necessitate the source database to be in read-only mode, and third-party tools lack native optimizations, complicating the process and increasing provisioning costs.
Innovation Solution
Implementing a rules engine that automates the selection and deployment of pluggable databases into suitable container databases using template pluggable databases and predicate rules, allowing for efficient cloning and relocation with minimal disruption and reduced costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the source PDB is placed in read-only mode for cloning operations, then cloning can be performed, but significant downtime occurs and write operations are prohibited
Solution Approach 1:
The patent implements preliminary actions by creating a read replica of the source PDB before the cloning operation begins. This read replica is prepared in advance and can be used immediately for cloning purposes, eliminating the need to place the source PDB in read-only mode and thus avoiding downtime while maintaining cloning reliability
Solution Approach 2:
The patent uses copying by creating a read replica (a copy) of the source PDB that can be used for cloning operations. Instead of cloning directly from the source PDB, the system copies the source data to a replica, which then serves as the cloning source, allowing the original source PDB to remain in read-write mode without downtime
2Adaptability or versatility
If third-party tools are used for cloning operations, then cloning capability is achieved, but provisioning costs increase and native optimizations are unavailable
Solution Approach 1:
The patent implements self-service by enabling the database system to perform cloning operations using its own native capabilities through read replicas. The system serves its own cloning needs internally without requiring external third-party tools, thereby eliminating additional provisioning costs while maintaining full cloning capability and utilizing native optimizations
Solution Approach 2:
The patent applies universality by making the read replica mechanism serve multiple functions: it acts as both a high-availability standby and a cloning source. This multi-functional approach eliminates the need for separate third-party cloning tools, reducing provisioning costs while maintaining versatile cloning capabilities across different scenarios
3Adaptability or versatility
If PDB files are moved between container databases, then relocation is achieved, but significant time is required especially for large PDBs
Solution Approach 1:
The patent uses copying by creating a read replica of the source PDB in the target container database. Instead of moving the actual PDB files which takes significant time, the system copies the data to create a replica in the target location, which can then be plugged in much faster, dramatically reducing relocation time while maintaining relocation capability
Solution Approach 2:
The patent implements preliminary action by pre-establishing the read replica in the target container database before the actual relocation is needed. The replica is prepared in advance with all necessary data, so when relocation is required, the pre-prepared replica can be quickly plugged into the target container, minimizing relocation time while maintaining full relocation capability
Data Source
AI summary
Techniques herein use rules automation and template pluggable databases to facilitate deployment into container databases. In an embodiment, a system of computers loads rules into a rules engine. Each rule associates a predicate with suitable container databases. The system receives a request to install a target pluggable database. The rules engine detects satisfied rules whose predicates match the request. Based on the suitable container databases of the satisfied rules, the rules engine selects a particular container database. The system installs the target pluggable database into the particular container database. In an embodiment, a system of computers stores a plurality of template pluggable databases in a repository. The repository receives an installation request. Based on the installation request, the system selects a particular template pluggable database. The system installs the particular template pluggable database into a container database.