Distributed Database State Snapshots for Resource Management

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

In distributed cloud environments, managing and synchronizing resource state information efficiently is challenging due to increased processing time, network bandwidth usage, and potential inconsistencies or deadlocks when each entity manages its own resources or when resource state information is managed in a central location.

Innovation Solution

The solution involves maintaining snapshots of state information for resource entities within a distributed cloud service using an object store, represented as a directed acyclic graph. This allows for efficient tracking of resource changes and allocation modifications by creating new snapshots for each change request, enabling concurrent processing of multiple requests without waiting for physical resource modifications to complete.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If each entity manages its own resources individually, then resource management autonomy is improved, but processing time and network bandwidth usage increase

Engineering Contradiction:

Improveresource management autonomyVSAvoidprocessing time

Core Design Contradiction:

Adaptability or versatilityVSLoss of time

Solution Approach 1:

The patent creates a snapshot (copy) of the resource state information and stores it in an object store. This copy allows the placement service to process change requests against the snapshot without blocking other operations. The snapshot is a directed acyclic graph that represents the state of resources at a specific point in time, enabling parallel processing while maintaining consistency.

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The system performs preliminary actions by creating and storing snapshots of resource state information before processing change requests. These snapshots are prepared in advance and can be used by multiple services simultaneously. The virtual lock mechanism is also established beforehand to manage access to snapshots, allowing services to acquire locks quickly without extensive coordination.

Inventive Principle:

Principle #10Preliminary action

2Adaptability or versatility

If each entity manages its own resources individually, then resource management autonomy is improved, but consistency and synchronization between entities deteriorate

Engineering Contradiction:

Improveresource management autonomyVSAvoidsynchronization consistency

Core Design Contradiction:

Adaptability or versatilityVSReliability

Solution Approach 1:

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The object store acts as an intermediary between the placement service and the resource management entities. It stores snapshots that serve as a common reference point for all services. The virtual lock mechanism also serves as an intermediary to coordinate access to snapshots, ensuring that services process requests in a consistent order without requiring continuous communication between entities.

Inventive Principle:

Principle #24Intermediary (Mediator)

3Reliability

If resource state information is managed in a central location, then consistency is improved, but service bottlenecks increase

Engineering Contradiction:

Improveresource data consistencyVSAvoidrequest processing throughput

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The patent segments the resource state information into multiple snapshots stored in an object store. Each snapshot represents a consistent state at a specific point in time and can be independently processed. This segmentation allows multiple services to work on different snapshots simultaneously, eliminating the bottleneck of a single centralized state management point while maintaining consistency within each snapshot.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

Inventive Principle:

Principle #10Preliminary action

4Reliability

If locking mechanisms are used to prevent resource competition, then resource allocation consistency is improved, but processing speed deteriorates due to extended locks

Engineering Contradiction:

Improveresource allocation consistencyVSAvoidrequest processing speed

Core Design Contradiction:

ReliabilityVSSpeed

Solution Approach 1:

Inventive Principle:

Principle #26Copying

Solution Approach 2:

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS12332871B2Representation and management of infrastructure state of a distributed database cloud service

Publication Date: 2025.06.17 ORACLE INT CORP

AI summary

Techniques are provided for maintaining snapshots of state information for a plurality of resource entities, within a distributed cloud service. A first snapshot of state information for the plurality of resource entities is maintained at an endpoint in persistent storage. A request to modify resources allocated to resource entities is received. A virtual lock on the state information reflected in the first snapshot is obtained. Upon obtaining the virtual lock, a service determines, based on the first snapshot, that there are available resources to perform the request to modify resources. A second snapshot of the state information, reflecting the modification of resources allocated to resource entities, is then generated and stored at the endpoint in persistent storage. The virtual lock on the state information is released and the resources allocated to the resource entities are modified according to the request.