Container Runtime State Checkpointing via Segmented Repository

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current methods lack the capability to effectively checkpoint, restore, store, manage, and share the runtime state of containers, which is crucial for maintaining the operational integrity of applications across different hosts and environments.

Innovation Solution

A method that captures a runtime state image of a container, including memory and file system state data, and stores it in a repository, allowing for restoration on the same or different physical hosts, while enabling versioning and sharing of container states through a common repository.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If container runtime state is captured and stored in a repository, then the ability to restore and share container states is improved, but the storage requirements and system complexity increase

Engineering Contradiction:

Improvecontainer runtime state restoration capabilityVSAvoidstate management system complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The container runtime state is segmented into distinct components: memory state data, file system state data, and process state data. Each component is captured, stored, and managed separately in the repository, allowing for selective restoration and reducing the complexity of managing monolithic state data.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

A repository acts as an intermediary between container instances and their runtime state data. The repository stores runtime state images and provides mechanisms for retrieving and restoring states, simplifying the complexity of direct state management between containers and their data.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Manufacturing precision

If complete runtime state images including memory and file system data are captured, then the restoration accuracy is improved, but the storage space requirements increase

Engineering Contradiction:

Improveruntime state restoration accuracyVSAvoidstorage space for state data

Core Design Contradiction:

Manufacturing precisionVSQuantity of substance

Solution Approach 1:

Only the necessary runtime state data required for container restoration is extracted and stored, rather than capturing entire container images. The system extracts memory state data, file system state data, and process state data specifically needed for state restoration, reducing storage requirements while maintaining restoration accuracy.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The system captures complete runtime state information needed for accurate restoration, but allows for selective restoration where only portions of the state data need to be retrieved. The repository stores complete state images but enables partial restoration based on specific needs, balancing storage requirements with restoration completeness.

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS11042450B2Mechanism for managing container runtime state

Publication Date: 2021.06.22 INTERNATIONAL BUSINESS MACHINE CORPORATION

AI summary

A method for checkpointing includes capturing a runtime state image of an application running in a container, wherein the runtime state image includes memory state data and file system state data of the container, adding the runtime state image to a collection of runtime state images associated with the container, and restoring the container to a specified state by retrieving at least a portion of the memory state data and file system state data of the runtime state image from the collection of runtime state images.