Multi-cluster Database Management via Self-service Templates

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing virtual computing systems face challenges in database provisioning and management, particularly in creating and registering databases, which are time-consuming and require interaction with multiple entities, and snapshot-based data protection methods limit the ability to recreate intermediate states, leading to potential data loss.

Innovation Solution

A database management system that allows users to quickly create and register databases through an intuitive interface with standardized configurations, and employs a combination of snapshots and transactional logs for comprehensive data protection, enabling point-in-time database recreation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If traditional database provisioning methods are used, then databases can be created with custom configurations, but the process is time-consuming and requires interaction with multiple entities

Engineering Contradiction:
Improvedatabase creation speedVSAvoidprovisioning process complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The system pre-configures standardized database templates with common configurations already established. Users can select from pre-defined templates that include database type, storage settings, security parameters, and other configurations, eliminating the need to manually configure each parameter from scratch and reducing both time and complexity.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system enables users to autonomously provision databases through a self-service interface without requiring interaction with multiple support entities. Users can independently select templates, configure parameters within allowed ranges, and deploy databases directly through the platform, streamlining the provisioning process.

Inventive Principle:
Principle #25Self-service

2Reliability

If snapshot-based data protection is used, then data can be protected at specific points in time, but the ability to recreate intermediate states is limited leading to potential data loss

Engineering Contradiction:
Improvedata protection capabilityVSAvoidintermediate data state loss
Core Design Contradiction:
ReliabilityVSLoss of information

Solution Approach 1:

The system continuously captures transactional logs between snapshots, maintaining an ongoing record of all data changes. This continuous logging ensures that no intermediate data states are lost, as the transactional logs provide a complete audit trail of all modifications that occur between snapshot points, enabling reconstruction of any intermediate state.

Inventive Principle:
Principle #20Continuity of useful action

Solution Approach 2:

The system combines snapshot technology with transactional logging to create a hybrid data protection approach. Snapshots provide periodic full or incremental backups, while transactional logs capture all intermediate changes. Together, they form a comprehensive protection mechanism that preserves both checkpoint states and intermediate modifications, eliminating the limitations of snapshot-only approaches.

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentEP4209924B1Multi-cluster database management system
Publication Date: 2025.12.17 NUTANIX INC
  • EP4209924B1 patent drawingFigure 1
  • EP4209924B1 patent drawingFigure 2
  • EP4209924B1 patent drawingFigure 3

AI summary

A multi-cluster database management system includes a first cluster in a database management system of a virtual computing system, a second cluster in the database management system, a server on the first cluster, a first agent on the first cluster, and a second agent on the second cluster. The server provides a database management service to a first database stored on the first cluster via the first agent and to a second database stored on the second cluster via the second agent.