Storage Device Recovery via Pseudo Boot Up

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Storage devices, such as SSDs and HDDs, enter a failure mode when operational parameters or metadata drift out of specified ranges, leading to data loss and inability to recover normal operation without re-initialization, which erases historical metadata and complicates device qualification.

Innovation Solution

A method and device configuration that allows for a pseudo boot up process, where the storage device restores to a prior operational state by switching from boot up to pseudo boot up using debug data like system metadata, mapping tables, and die erase time, enabling operation resumption without losing historical data.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If the storage device enters failure mode and is re-initialized to recover normal operation, then the device can return to normal operation, but the historical metadata and device lifetime information are lost

Engineering Contradiction:
Improvedevice operation recoveryVSAvoidhistorical metadata
Core Design Contradiction:
ReliabilityVSLoss of information

Solution Approach 1:

The patent implements a preliminary action by creating a checkpoint or snapshot of the file system state before the device enters failure mode. This checkpoint contains critical metadata and device state information that can be restored later, preventing complete data loss when re-initialization is necessary to recover from failure mode.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent creates a copy of the essential metadata and device state information in a protected format before failure occurs. This copy serves as a backup that can be used to restore the device to its pre-failure state, preserving historical information while enabling recovery from failure mode.

Inventive Principle:
Principle #26Copying

2Reliability

If journaling is implemented to recover device state, then device recovery is possible, but the journaling process is complex and often fails to replay correctly

Engineering Contradiction:
Improvedevice recovery capabilityVSAvoidjournaling process
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent extracts only the essential critical metadata and device state information needed for recovery, rather than implementing a comprehensive journaling system that logs all operations. This selective extraction simplifies the recovery process while maintaining effectiveness, avoiding the complexity and failure modes of full journaling implementation.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

Instead of using journaling to reconstruct device state forward from a baseline, the patent inverts the approach by capturing the complete device state at checkpoint moments and restoring directly from these snapshots. This reversal eliminates the complex replay process inherent in traditional journaling.

Inventive Principle:
Principle #13The other way round (Inversion)

3Loss of information

If the device operates continuously without re-initialization, then historical metadata is preserved, but the device cannot recover from failure mode when operational parameters drift

Engineering Contradiction:
Improvehistorical metadata preservationVSAvoidfailure mode recovery
Core Design Contradiction:
Loss of informationVSReliability

Solution Approach 1:

The patent implements preliminary action by establishing checkpoints of device state before failure mode occurs. These checkpoints allow the device to be restored to a known good state without complete re-initialization, thus preserving historical metadata while enabling failure recovery.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent changes the recovery parameter from complete re-initialization (which loses all historical data) to selective restoration from checkpoints. This parameter change allows recovery from failure mode while maintaining preservation of historical metadata that occurred after the checkpoint was created.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS11928036B2Recovery from broken mode
Publication Date: 2024.03.12 KIOXIA CORP
  • US11928036B2 patent drawing
  • US11928036B2 patent drawing
  • US11928036B2 patent drawing

AI summary

A method of recovering a storage device from a failure mode by a controller, the method comprising executing a boot up of the storage device, determining whether an instruction for pseudo boot up of the storage device has been received, switching from the boot up to a pseudo boot up of the storage device if an instruction for the pseudo boot up has been received, and resuming operation of the storage device from the prior operational state. The pseudo boot up restores the storage device to a prior operational state up to the point in time when the storage device encountered a failure triggering event that caused the storage device to enter the failure mode.