Stateful Application Snapshotting for Reclaimable Compute Instances

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Reclaimable compute instances pose challenges for stateful applications due to unpredictable termination, leading to loss of application state data and resulting in issues like data inconsistency, poor user experience, and reduced fault tolerance.

Innovation Solution

A resource management service predicts the remaining lifespan of reclaimable compute instances and, when it falls below a threshold, creates an application state snapshot, which is stored remotely, allowing the application to be restored on a new instance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of energy

If reclaimable compute instances are used to reduce costs, then computing resource cost is reduced, but application state data is lost when instances are terminated

Engineering Contradiction:

Improvecomputing resource costVSAvoidapplication state data

Core Design Contradiction:

Loss of energyVSLoss of information

Solution Approach 1:

The system performs preliminary actions by continuously monitoring predicted remaining lifespans of reclaimable compute instances and proactively creating application state snapshots before termination occurs. When the predicted lifespan falls below a threshold, the system captures the application state in advance, ensuring data preservation before the instance is reclaimed.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system creates copies of the application state data by generating snapshots of memory contents and storing them in persistent storage. These snapshots serve as replicas that can be restored to new compute instances, allowing the application state to be preserved and transferred without losing information when the original instance is terminated.

Inventive Principle:

Principle #26Copying

2Loss of energy

If reclaimable compute instances are used, then cost savings are achieved, but computational downtime occurs during instance termination and reconnection

Engineering Contradiction:

Improvecomputing resource costVSAvoidcomputational downtime

Core Design Contradiction:

Loss of energyVSLoss of time

Solution Approach 1:

The system performs preliminary restoration by loading application state snapshots into memory of new compute instances before the application needs to resume execution. This pre-positioning of state data eliminates waiting time during instance transitions, allowing the application to continue with minimal or no perceptible downtime.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system maintains continuity of useful action by ensuring the application state is preserved and rapidly restored across instance transitions. Through continuous monitoring, proactive snapshot creation, and expedited restoration processes, the system minimizes interruptions and keeps the application operational throughout the instance lifecycle.

Inventive Principle:

Principle #20Continuity of useful action

3Productivity

If application state is maintained in memory on reclaimable instances, then application performance is improved, but data consistency is lost when instances are reclaimed

Engineering Contradiction:

Improveapplication performanceVSAvoiddata consistency

Core Design Contradiction:

ProductivityVSStability of the object's composition

Solution Approach 1:

The system creates accurate copies of the application state by snapshotting memory contents at specific points in time. These snapshots capture the consistent state of the application, including all data structures and variables, allowing the application to resume from a known good state on new instances without data inconsistency.

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The system implements feedback mechanisms by continuously monitoring the predicted remaining lifespan of compute instances and comparing it against threshold values. This feedback loop triggers proactive snapshot creation when termination is anticipated, ensuring data consistency is maintained through timely state preservation before instance reclamation.

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS12572428B2Stateful application persistency for reclaimable compute instances

Publication Date: 2026.03.10 NETAPP INC

AI summary

Disclosed herein are methods and systems for the operation of a resource management service. The resource management service deploys reclaimable compute instances from a resource pool and continuously generates predicted remaining lifespans for the deployed reclaimable compute instances. The predicted remaining lifespan is monitored to determine if the predicted remaining lifespan is below a threshold value. In response to the predicted remaining lifespan for a reclaimable compute instance falling below a threshold value, the resource management service instructs the reclaimable compute instance to create an application state snapshot of an application running thereon. A subsequent compute instance is deployed from the resource pool, on which the application can be restored to a previous state using the application state snapshot.