Proactive Snapshot Generation Using ML Prediction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current snapshot generation systems are reactive and computationally expensive, often generating redundant snapshots that fail to capture the 'good state' before an event, and are inefficient in managing and organizing stored snapshots for effective recovery.

Innovation Solution

A system that uses application-specific triggering criteria and machine learning models to predict when an event of interest will occur, generating snapshots proactively and storing them in a hybrid cloud network, along with a snapshot database that aggregates state and feature information for efficient snapshot management and recovery.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If snapshots are generated reactively based on events, then snapshots capture actual system states, but the system is computationally expensive and generates redundant snapshots

Engineering Contradiction:

Improvesnapshot accuracyVSAvoidsnapshot generation efficiency

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The system performs preliminary actions by proactively generating snapshots before events occur based on predicted event timing. The machine learning model forecasts when events will happen, and snapshots are taken in advance during normal operation, avoiding the computational overhead of reactive snapshot generation during actual events.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system uses feedback from machine learning models that analyze system state patterns to determine when proactive snapshots should be generated. The model learns from historical data about system behavior and provides feedback on optimal snapshot timing, reducing redundant snapshots while maintaining reliability.

Inventive Principle:

Principle #23Feedback

2Productivity

If snapshots are generated proactively using machine learning predictions, then redundancy is reduced and computational efficiency improves, but the system complexity increases

Engineering Contradiction:

Improvesnapshot generation efficiencyVSAvoidsystem complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The machine learning model acts as an intermediary between the distributed system and the snapshot generation mechanism. It processes system state data and provides predictions about event timing, which then guide the snapshot generation process, simplifying the overall control logic while maintaining high efficiency.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The system replaces manual or rule-based snapshot generation mechanisms with machine learning-based predictive mechanisms. This substitution automates the decision-making process for snapshot generation, reducing the need for complex manual configuration and management while improving efficiency.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Ease of manufacture

If snapshots are taken at regular intervals, then snapshot generation is simple and consistent, but redundant snapshots are generated that do not capture meaningful state changes

Engineering Contradiction:

Improvesnapshot generation simplicityVSAvoiddata redundancy

Core Design Contradiction:

Ease of manufactureVSLoss of substance

Solution Approach 1:

The system dynamically changes the snapshot generation parameters based on predicted event timing. Instead of fixed regular intervals, the system adjusts snapshot frequency and timing according to ML predictions of when events will occur, reducing redundancy while maintaining simplicity in the generation process.

Inventive Principle:

Principle #35Parameter changes

4Productivity

If a snapshot database aggregates state and feature information from multiple nodes, then snapshot management and recovery efficiency improve, but the quantity of data to be managed increases

Engineering Contradiction:

Improverecovery efficiencyVSAvoiddata volume

Core Design Contradiction:

ProductivityVSQuantity of substance

Solution Approach 1:

The system extracts only the essential state and feature information needed for recovery from multiple distributed nodes and aggregates it in a centralized snapshot database. This selective extraction and aggregation reduces the total data volume compared to storing complete node states, while maintaining recovery efficiency.

Inventive Principle:

Principle #2Taking out (Extraction)

Data Source

PatentUS11467922B2Intelligent snapshot generation and recovery in a distributed system

Publication Date: 2022.10.11 CISCO TECHNOLOGY INC

AI summary

Systems, methods, and storage media for generating and/or recoverying an end-to-end hybrid cloud snapshot are disclosed. Exemplary implementations may: receive application specific triggering criteria for generating an end-to-end hybrid cloud snapshot; trigger a snapshot event based on the application specific triggering criteria and a machine learning model; accumulate snapshot portions from a plurality of nodes in a hybrid cloud network; and store, in a snapshot database, the end-to-end hybrid cloud snapshot comprising the snapshot portions from the plurality of nodes. Exemplary implementations may further include: generate a recovery model based on a snapshot database, wherein the snapshot database comprises a plurality of network snapshots associated with a network, receive parameters for a desired recovery state, and identify a recommended network snapshot from the plurality of network snapshots based on the recovery model and the parameters for the desired recovery state.