Recovery Agents for Computing System Component Restoration

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing computing systems face disruptions and loss of diagnostic information when attempting to recover from system errors or security threats, as shutting down the entire system can be time-consuming and destructive, especially when the issue is confined to a discrete component, and valuable data is lost in the process.

Innovation Solution

A computing system architecture with multiple abstraction layers, each equipped with recovery agents that receive recovery scripts from an external repository, allowing for targeted capture and recovery of compromised components without disrupting the entire system, using recovery scripts that include actions to send snapshots to a forensic repository or restore components to a known good state.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If the entire system is shut down to recover from system errors or security threats, then system integrity is restored, but valuable diagnostic information is lost and system disruption increases

Engineering Contradiction:
Improvesystem integrityVSAvoiddiagnostic information
Core Design Contradiction:
ReliabilityVSLoss of information

Solution Approach 1:

The system is divided into multiple discrete recoverable components (virtual machines, containers, processes) that can be individually captured and restored. Recovery agents are deployed at different abstraction layers (host, VM, container) to enable granular recovery operations without requiring complete system shutdown, thus preserving diagnostic information while restoring integrity.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Forensic snapshots and diagnostic data are captured before recovery actions are executed. Recovery agents proactively collect system state information, memory dumps, and log data prior to restoring components to known good states, ensuring diagnostic information is preserved even as the system recovers from compromised components.

Inventive Principle:
Principle #10Preliminary action

2Reliability

If the entire system is shut down for recovery, then system security is improved, but recovery time increases

Engineering Contradiction:
Improvesystem securityVSAvoidrecovery time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

Recovery operations are performed on individual components (containers, virtual machines, processes) rather than the entire system. This allows non-critical components to continue operating during recovery, significantly reducing overall system downtime while maintaining security through targeted restoration of compromised elements.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Recovery agents are automatically deployed and executed without requiring manual intervention or complete system shutdown. The system performs self-recovery by automatically capturing forensic data, identifying compromised components, and restoring them to known good states, thereby reducing both recovery time and operational disruption.

Inventive Principle:
Principle #25Self-service

3Productivity

If recovery actions are performed on discrete components, then system disruption is minimized, but system complexity increases

Engineering Contradiction:
Improvesystem availabilityVSAvoidsystem architecture
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

A unified recovery service architecture is implemented that operates across multiple abstraction layers (host, virtual machine, container). The same recovery service framework and agent deployment mechanism handles recovery at different levels, providing a universal solution that manages complexity through standardization while enabling granular recovery operations.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

Recovery agents serve as intermediaries between the recovery service and various system components. These agents abstract the complexity of component-specific recovery operations, providing a standardized interface that simplifies the overall system architecture while enabling precise control over discrete component recovery.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP3370151B1Recovery services for computing systems
Publication Date: 2019.11.13 HEWLETT PACKARD ENTERPRISE DEV LP
  • EP3370151B1 patent drawingFigure 1A
  • EP3370151B1 patent drawingFigure 1B
  • EP3370151B1 patent drawingFigure 2

AI summary

Example implementations relate to capturing and/or recovering components of a computing system. A recovery service may receive a recovery script from an external recovery script repository, wherein the recovery script may include a number of actions, each respective action being a capture action or a recovery action. For action in the recovery script, the recovery service may request a recovery agent to perform the action on a component of the computing system.